{
    "metadata": {
        "benchmark": "Legal Research Bench",
        "slug": "legal_research",
        "description": "Evaluating agents on legal research tasks across diverse areas of US law",
        "benchmark_id": "legal_research",
        "family": "legal_research",
        "version": "1",
        "updated": "2026-11-05",
        "dataset_type": "private",
        "industry": "legal",
        "tasks": {
            "overall": "Overall",
            "administrative_regulatory": "Administrative / Regulatory",
            "business_commercial": "Business & Commercial",
            "civil_litigation": "Civil Litigation",
            "constitutional_civil_rights": "Constitutional / Civil Rights",
            "criminal": "Criminal",
            "family": "Family",
            "health": "Health",
            "immigration": "Immigration",
            "weighted_pass": "Weighted Pass",
            "weighted_pass__administrative_regulatory": "Administrative / Regulatory",
            "weighted_pass__business_commercial": "Business & Commercial",
            "weighted_pass__civil_litigation": "Civil Litigation",
            "weighted_pass__constitutional_civil_rights": "Constitutional / Civil Rights",
            "weighted_pass__criminal": "Criminal",
            "weighted_pass__family": "Family",
            "weighted_pass__health": "Health",
            "weighted_pass__immigration": "Immigration"
        },
        "models": [
            "alibaba/qwen3.6-plus",
            "alibaba/qwen3.7-max",
            "alibaba/qwen3.7-plus",
            "alibaba/qwen3.8-27b",
            "alibaba/qwen3.8-max",
            "alibaba/qwen3.8-max-internal",
            "ant/ling-3.0-flash-2607",
            "ant/ling-3.0-flash-af-rc3",
            "anthropic/claude-fable-5",
            "anthropic/claude-fable-5-1",
            "anthropic/claude-haiku-4-5-20251001-thinking",
            "anthropic/claude-haiku-5-5",
            "anthropic/claude-opus-4-7",
            "anthropic/claude-opus-4-8",
            "anthropic/claude-opus-5",
            "anthropic/claude-opus-5-5",
            "anthropic/claude-sonnet-4-6",
            "anthropic/claude-sonnet-5",
            "anthropic/claude-sonnet-5-5",
            "baseten/nemotron-3.5-super-vl-120b-a12b",
            "deepseek/deepseek-v4-flash-0731",
            "deepseek/deepseek-v4-pro",
            "deepseek/deepseek-v4-pro-0813",
            "deepseek/deepseek-v4.1-flash",
            "deepseek/deepseek-v4.1-flash-1017",
            "fireworks/ember-1",
            "fireworks/nemotron-lightning-3p5-30b-a3b",
            "google/gemini-3-flash-preview",
            "google/gemini-3.1-flash-lite-preview",
            "google/gemini-3.1-pro-preview",
            "google/gemini-3.5-flash",
            "google/gemini-3.5-flash-lite",
            "google/gemini-3.6-flash",
            "google/gemini-3.7-flash",
            "google/gemini-3.8-flash",
            "google/gemini-4-argon",
            "google/gemini-4-carbon",
            "grok/grok-4.20-0309-reasoning",
            "grok/grok-4.3",
            "grok/grok-4.5",
            "grok/grok-4.6",
            "grok/grok-4.7",
            "inception/mercury-2.5",
            "inception/mercury-3",
            "kimi/kimi-k2.5-thinking",
            "kimi/kimi-k2.6",
            "kimi/kimi-k3",
            "kimi/kimi-k3.1-pro",
            "meta/muse-spark-open",
            "meta/muse-wonder",
            "meta/muse-wonder-full",
            "meta/muse_spark_1_1",
            "meta/muse_spark_1_2",
            "meta/muse_spark_1_3",
            "meta/muse_spark_1_3_max",
            "minimax/MiniMax-M2.7",
            "minimax/MiniMax-M3",
            "minimax/MiniMax-M3.1-flash",
            "mistralai/mistral-large-4",
            "mistralai/mistral-large-4-oss",
            "mistralai/mistral-medium-3.5",
            "nvidia/nemotron-3-ultra-550b-a55b",
            "openai/gpt-5.4-mini-2026-03-17",
            "openai/gpt-5.4-nano-2026-03-17",
            "openai/gpt-5.5",
            "openai/gpt-5.6-luna",
            "openai/gpt-5.6-sol",
            "openai/gpt-5.6-terra",
            "openai/gpt-6-astra",
            "openai/gpt-6-luna",
            "openai/gpt-6-sol",
            "openai/gpt-6.1-sol",
            "openai/gpt-6.1-terra",
            "poolside/laguna-m.1",
            "poolside/laguna-xs.2",
            "reflection/Beam-1.0",
            "stepfun/step-5-preview",
            "tencent/hy4",
            "tencent/hy4-preview",
            "thinkingmachines/inkling",
            "thinkingmachines/inkling-2-small",
            "thinkingmachines/inkling-small",
            "xiaomi/mimo-v2.5",
            "xiaomi/mimo-v2.5-pro",
            "xiaomi/mimo-v2.6-flash",
            "xiaomi/mimo-v2.6-pro",
            "zai/glm-5.1",
            "zai/glm-5.2",
            "zai/glm-5.3",
            "zai/glm-5.3-flash",
            "zai/glm-legal"
        ],
        "partners": [
            {
                "name": "fisher_phillips",
                "url": "https://www.fisherphillips.com/"
            },
            {
                "name": "mcdermott",
                "url": "https://www.mcdermottlaw.com/"
            },
            {
                "name": "reed_smith",
                "url": "https://www.reedsmith.com/"
            },
            {
                "name": "legaltech_hub",
                "url": "https://www.legaltechnologyhub.com/"
            }
        ],
        "showBadge": false,
        "visible": true,
        "use_cost_per_test": true,
        "token_tasks_measured": 208,
        "runner": "external",
        "mode": "agentic",
        "archived": false,
        "partner": false,
        "total_models": 91
    },
    "tasks": {
        "overall": {
            "google/gemini-4-carbon": {
                "accuracy": 55.769,
                "latency": 3112.579,
                "stderr": 3.452,
                "cost_per_test": 8.743524,
                "token_totals": {
                    "input_tokens": 1062929092,
                    "output_tokens": 52772544,
                    "reasoning_tokens": 36254828,
                    "cache_read_tokens": 918030058,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 39.2,
                    "avg_query_seconds": 79.5,
                    "median_query_seconds": 33.5,
                    "avg_ttft_seconds": 12.7,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 55.288,
                "latency": 317.461,
                "stderr": 3.456,
                "cost_per_test": 0.588812,
                "token_totals": {
                    "input_tokens": 240995118,
                    "output_tokens": 10646285,
                    "reasoning_tokens": 6279342,
                    "cache_read_tokens": 205258324,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 205,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 25.0,
                    "avg_query_seconds": 12.5,
                    "median_query_seconds": 4.6,
                    "avg_ttft_seconds": 4.9,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 55.288,
                "latency": 1860.719,
                "stderr": 3.456,
                "cost_per_test": 6.762498,
                "token_totals": {
                    "input_tokens": 626960894,
                    "output_tokens": 23871466,
                    "reasoning_tokens": 9873124,
                    "cache_read_tokens": 540621000,
                    "cache_write_tokens": 86242293
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 37.0,
                    "avg_query_seconds": 50.3,
                    "median_query_seconds": 27.5,
                    "avg_ttft_seconds": 14.4,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 55.288,
                "latency": 3449.382,
                "stderr": 3.456,
                "cost_per_test": 23.055789,
                "token_totals": {
                    "input_tokens": 1085827683,
                    "output_tokens": 50734245,
                    "reasoning_tokens": 31513293,
                    "cache_read_tokens": 927334456,
                    "cache_write_tokens": 158405763
                },
                "token_tasks_measured": 206,
                "turn_metrics": {
                    "tasks_measured": 206,
                    "avg_turns": 48.5,
                    "avg_query_seconds": 71.1,
                    "median_query_seconds": 16.8,
                    "avg_ttft_seconds": 14.4,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 54.808,
                "latency": 1881.045,
                "stderr": 3.459,
                "cost_per_test": 6.44908,
                "token_totals": {
                    "input_tokens": 915877177,
                    "output_tokens": 24432059,
                    "reasoning_tokens": 11886477,
                    "cache_read_tokens": 739668768,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 38.2,
                    "avg_query_seconds": 49.2,
                    "median_query_seconds": 16.4,
                    "avg_ttft_seconds": 8.1,
                    "output_time_split": {
                        "reasoning_pct": 79.0,
                        "content_pct": 3.0,
                        "tool_call_pct": 18.1
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 53.365,
                "latency": 1654.378,
                "stderr": 3.467,
                "cost_per_test": 3.595401,
                "token_totals": {
                    "input_tokens": 760777228,
                    "output_tokens": 19689738,
                    "reasoning_tokens": 13012367,
                    "cache_read_tokens": 707647460,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 45.3,
                    "avg_query_seconds": 36.5,
                    "median_query_seconds": 9.9,
                    "avg_ttft_seconds": 8.2,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 50.481,
                "latency": 927.111,
                "stderr": 3.475,
                "cost_per_test": 3.980242,
                "token_totals": {
                    "input_tokens": 732491974,
                    "output_tokens": 17268053,
                    "reasoning_tokens": 11668590,
                    "cache_read_tokens": 644062808,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 46.7,
                    "avg_query_seconds": 19.9,
                    "median_query_seconds": 7.2,
                    "avg_ttft_seconds": 5.5,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 50.481,
                "latency": 6749.531,
                "stderr": 3.475,
                "cost_per_test": 23.613691,
                "token_totals": {
                    "input_tokens": 1686854010,
                    "output_tokens": 141554772,
                    "reasoning_tokens": 123959393,
                    "cache_read_tokens": 1436953409,
                    "cache_write_tokens": 249809865
                },
                "token_tasks_measured": 185,
                "turn_metrics": {
                    "tasks_measured": 197,
                    "avg_turns": 72.9,
                    "avg_query_seconds": 89.0,
                    "median_query_seconds": 9.2,
                    "avg_ttft_seconds": 9.7,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 49.519,
                "latency": 1337.604,
                "stderr": 3.475,
                "cost_per_test": 9.791426,
                "token_totals": {
                    "input_tokens": 210052964,
                    "output_tokens": 22537892,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 150055095,
                    "cache_write_tokens": 59958741
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 14.6,
                    "avg_query_seconds": 91.5,
                    "median_query_seconds": 56.5,
                    "avg_ttft_seconds": 37.0,
                    "output_time_split": {
                        "reasoning_pct": 0.0,
                        "content_pct": 0.8,
                        "tool_call_pct": 99.2
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 49.038,
                "latency": 3110.171,
                "stderr": 3.475,
                "cost_per_test": 2.241388,
                "token_totals": {
                    "input_tokens": 458279584,
                    "output_tokens": 55494507,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 368033787,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 23.3,
                    "avg_query_seconds": 133.6,
                    "median_query_seconds": 67.1,
                    "avg_ttft_seconds": 3.4,
                    "output_time_split": {
                        "reasoning_pct": 82.2,
                        "content_pct": 1.1,
                        "tool_call_pct": 16.7
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 48.077,
                "latency": 1504.077,
                "stderr": 3.473,
                "cost_per_test": 1.530204,
                "token_totals": {
                    "input_tokens": 274776010,
                    "output_tokens": 12377024,
                    "reasoning_tokens": 7368696,
                    "cache_read_tokens": 203687808,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 20.9,
                    "avg_query_seconds": 72.1,
                    "median_query_seconds": 34.7,
                    "avg_ttft_seconds": 17.7,
                    "output_time_split": null
                },
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 48.077,
                "latency": 4631.383,
                "stderr": 3.473,
                "cost_per_test": 21.605566,
                "token_totals": {
                    "input_tokens": 1487117357,
                    "output_tokens": 70127005,
                    "reasoning_tokens": 61207392,
                    "cache_read_tokens": 1241425920,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 47.3,
                    "avg_query_seconds": 97.8,
                    "median_query_seconds": 10.6,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 48.077,
                "latency": 5588.989,
                "stderr": 3.473,
                "cost_per_test": 13.663179,
                "token_totals": {
                    "input_tokens": 2650038387,
                    "output_tokens": 173348390,
                    "reasoning_tokens": 151205430,
                    "cache_read_tokens": 2338418360,
                    "cache_write_tokens": 311466299
                },
                "token_tasks_measured": 201,
                "turn_metrics": {
                    "tasks_measured": 204,
                    "avg_turns": 85.2,
                    "avg_query_seconds": 64.6,
                    "median_query_seconds": 11.2,
                    "avg_ttft_seconds": 8.2,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 47.596,
                "latency": 3677.636,
                "stderr": 3.471,
                "cost_per_test": 2.486351,
                "token_totals": {
                    "input_tokens": 544286180,
                    "output_tokens": 42522887,
                    "reasoning_tokens": 35664649,
                    "cache_read_tokens": 472313472,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 35.5,
                    "avg_query_seconds": 103.5,
                    "median_query_seconds": 58.3,
                    "avg_ttft_seconds": 27.2,
                    "output_time_split": {
                        "reasoning_pct": 76.7,
                        "content_pct": 1.1,
                        "tool_call_pct": 22.1
                    }
                },
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 47.115,
                "latency": 1274.337,
                "stderr": 3.469,
                "cost_per_test": 4.920812,
                "token_totals": {
                    "input_tokens": 476673717,
                    "output_tokens": 14028919,
                    "reasoning_tokens": 9121242,
                    "cache_read_tokens": 350504320,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 44.0,
                    "avg_query_seconds": 28.9,
                    "median_query_seconds": 12.5,
                    "avg_ttft_seconds": 3.3,
                    "output_time_split": {
                        "reasoning_pct": 78.7,
                        "content_pct": 12.9,
                        "tool_call_pct": 8.5
                    }
                },
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 47.115,
                "latency": 1816.322,
                "stderr": 3.469,
                "cost_per_test": 0.181201,
                "token_totals": {
                    "input_tokens": 246448526,
                    "output_tokens": 20667764,
                    "reasoning_tokens": 15354087,
                    "cache_read_tokens": 204079360,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 205,
                "turn_metrics": {
                    "tasks_measured": 206,
                    "avg_turns": 21.3,
                    "avg_query_seconds": 85.4,
                    "median_query_seconds": 43.3,
                    "avg_ttft_seconds": 5.2,
                    "output_time_split": {
                        "reasoning_pct": 81.4,
                        "content_pct": 1.9,
                        "tool_call_pct": 16.6
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 46.635,
                "latency": 735.191,
                "stderr": 3.467,
                "cost_per_test": 0.288586,
                "token_totals": {
                    "input_tokens": 701422864,
                    "output_tokens": 24259680,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 646891008,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 48.5,
                    "avg_query_seconds": 15.1,
                    "median_query_seconds": 4.4,
                    "avg_ttft_seconds": 1.8,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 46.154,
                "latency": 760.122,
                "stderr": 3.465,
                "cost_per_test": 3.469601,
                "token_totals": {
                    "input_tokens": 672994446,
                    "output_tokens": 6248749,
                    "reasoning_tokens": 3873135,
                    "cache_read_tokens": 651592955,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 37.2,
                    "avg_query_seconds": 20.4,
                    "median_query_seconds": 10.2,
                    "avg_ttft_seconds": 4.2,
                    "output_time_split": {
                        "reasoning_pct": 61.9,
                        "content_pct": 2.6,
                        "tool_call_pct": 35.5
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 46.154,
                "latency": 7649.584,
                "stderr": 3.465,
                "cost_per_test": 5.312295,
                "token_totals": {
                    "input_tokens": 1691431541,
                    "output_tokens": 63439847,
                    "reasoning_tokens": 53571199,
                    "cache_read_tokens": 1582915968,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 187,
                "turn_metrics": {
                    "tasks_measured": 187,
                    "avg_turns": 72.2,
                    "avg_query_seconds": 105.9,
                    "median_query_seconds": 44.3,
                    "avg_ttft_seconds": 21.5,
                    "output_time_split": {
                        "reasoning_pct": 81.4,
                        "content_pct": 0.2,
                        "tool_call_pct": 18.5
                    }
                },
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 45.192,
                "latency": 2106.881,
                "stderr": 3.459,
                "cost_per_test": 0.69856,
                "token_totals": {
                    "input_tokens": 2003160078,
                    "output_tokens": 87441264,
                    "reasoning_tokens": 76652436,
                    "cache_read_tokens": 1909100928,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 74.5,
                    "avg_query_seconds": 28.2,
                    "median_query_seconds": 6.4,
                    "avg_ttft_seconds": 1.6,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 45.192,
                "latency": 3067.486,
                "stderr": 3.459,
                "cost_per_test": 0.240392,
                "token_totals": {
                    "input_tokens": 636228822,
                    "output_tokens": 43453265,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 559661440,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 30.9,
                    "avg_query_seconds": 99.2,
                    "median_query_seconds": 40.1,
                    "avg_ttft_seconds": 1.8,
                    "output_time_split": {
                        "reasoning_pct": 85.5,
                        "content_pct": 0.9,
                        "tool_call_pct": 13.6
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 45.192,
                "latency": 3839.511,
                "stderr": 3.459,
                "cost_per_test": 0.846372,
                "token_totals": {
                    "input_tokens": 312749338,
                    "output_tokens": 50001075,
                    "reasoning_tokens": 43836379,
                    "cache_read_tokens": 264949824,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 31.0,
                    "avg_query_seconds": 123.9,
                    "median_query_seconds": 36.1,
                    "avg_ttft_seconds": 3.4,
                    "output_time_split": {
                        "reasoning_pct": 72.0,
                        "content_pct": 8.7,
                        "tool_call_pct": 19.3
                    }
                },
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 44.712,
                "latency": 2313.552,
                "stderr": 3.456,
                "cost_per_test": 3.674091,
                "token_totals": {
                    "input_tokens": 899320434,
                    "output_tokens": 18199298,
                    "reasoning_tokens": 11913573,
                    "cache_read_tokens": 817311062,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 40.2,
                    "avg_query_seconds": 57.5,
                    "median_query_seconds": 21.2,
                    "avg_ttft_seconds": 6.0,
                    "output_time_split": {
                        "reasoning_pct": 50.2,
                        "content_pct": 2.3,
                        "tool_call_pct": 47.5
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 43.75,
                "latency": 306.893,
                "stderr": 3.448,
                "cost_per_test": 0.508097,
                "token_totals": {
                    "input_tokens": 163340498,
                    "output_tokens": 5837394,
                    "reasoning_tokens": 2676881,
                    "cache_read_tokens": 113477050,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 205,
                "turn_metrics": {
                    "tasks_measured": 205,
                    "avg_turns": 18.1,
                    "avg_query_seconds": 17.0,
                    "median_query_seconds": 6.2,
                    "avg_ttft_seconds": 6.3,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 43.75,
                "latency": 714.036,
                "stderr": 3.448,
                "cost_per_test": 2.820942,
                "token_totals": {
                    "input_tokens": 136489421,
                    "output_tokens": 12617848,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 98212592,
                    "cache_write_tokens": 38178928
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 11.7,
                    "avg_query_seconds": 61.7,
                    "median_query_seconds": 48.1,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 43.75,
                "latency": 4979.806,
                "stderr": 3.448,
                "cost_per_test": 1.647359,
                "token_totals": {
                    "input_tokens": 1674824268,
                    "output_tokens": 253184033,
                    "reasoning_tokens": 237163492,
                    "cache_read_tokens": 1419547906,
                    "cache_write_tokens": 255154225
                },
                "token_tasks_measured": 206,
                "turn_metrics": {
                    "tasks_measured": 206,
                    "avg_turns": 60.8,
                    "avg_query_seconds": 81.9,
                    "median_query_seconds": 14.1,
                    "avg_ttft_seconds": 8.8,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 41.827,
                "latency": 1545.997,
                "stderr": 3.428,
                "cost_per_test": 2.719446,
                "token_totals": {
                    "input_tokens": 435277342,
                    "output_tokens": 29025915,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 353355756,
                    "cache_write_tokens": 81742641
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 24.7,
                    "avg_query_seconds": 62.6,
                    "median_query_seconds": 26.2,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 41.827,
                "latency": 2388.741,
                "stderr": 3.428,
                "cost_per_test": 0.04878,
                "token_totals": {
                    "input_tokens": 203583535,
                    "output_tokens": 5833302,
                    "reasoning_tokens": 4185422,
                    "cache_read_tokens": 177157760,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 22.1,
                    "avg_query_seconds": 108.1,
                    "median_query_seconds": 61.1,
                    "avg_ttft_seconds": 9.4,
                    "output_time_split": {
                        "reasoning_pct": 65.0,
                        "content_pct": 0.8,
                        "tool_call_pct": 34.2
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 41.346,
                "latency": 701.783,
                "stderr": 3.423,
                "cost_per_test": 0.24793,
                "token_totals": {
                    "input_tokens": 410443550,
                    "output_tokens": 29229774,
                    "reasoning_tokens": 25253265,
                    "cache_read_tokens": 362718592,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 26.1,
                    "avg_query_seconds": 26.9,
                    "median_query_seconds": 11.6,
                    "avg_ttft_seconds": 2.1,
                    "output_time_split": {
                        "reasoning_pct": 76.9,
                        "content_pct": 1.1,
                        "tool_call_pct": 22.0
                    }
                },
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 41.346,
                "latency": 4172.869,
                "stderr": 3.423,
                "cost_per_test": 7.907299,
                "token_totals": {
                    "input_tokens": 1570540015,
                    "output_tokens": 81682466,
                    "reasoning_tokens": 73169720,
                    "cache_read_tokens": 1375861858,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 46.6,
                    "avg_query_seconds": 89.6,
                    "median_query_seconds": 11.8,
                    "avg_ttft_seconds": 5.2,
                    "output_time_split": {
                        "reasoning_pct": 48.7,
                        "content_pct": 3.0,
                        "tool_call_pct": 48.3
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 41.346,
                "latency": 5455.029,
                "stderr": 3.423,
                "cost_per_test": 2.222804,
                "token_totals": {
                    "input_tokens": 1641853555,
                    "output_tokens": 106544013,
                    "reasoning_tokens": 94881685,
                    "cache_read_tokens": 1520894592,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 194,
                "turn_metrics": {
                    "tasks_measured": 202,
                    "avg_turns": 68.1,
                    "avg_query_seconds": 79.1,
                    "median_query_seconds": 9.9,
                    "avg_ttft_seconds": 3.8,
                    "output_time_split": null
                },
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 40.865,
                "latency": 530.813,
                "stderr": 3.417,
                "cost_per_test": 0.650063,
                "token_totals": {
                    "input_tokens": 193686912,
                    "output_tokens": 8979986,
                    "reasoning_tokens": 5305335,
                    "cache_read_tokens": 133646110,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 205,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 20.6,
                    "avg_query_seconds": 25.4,
                    "median_query_seconds": 9.5,
                    "avg_ttft_seconds": 4.5,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 40.865,
                "latency": 2527.177,
                "stderr": 3.417,
                "cost_per_test": 1.088017,
                "token_totals": {
                    "input_tokens": 433434003,
                    "output_tokens": 35408474,
                    "reasoning_tokens": 30213976,
                    "cache_read_tokens": 380911360,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 30.2,
                    "avg_query_seconds": 83.7,
                    "median_query_seconds": 27.7,
                    "avg_ttft_seconds": 2.4,
                    "output_time_split": {
                        "reasoning_pct": 76.2,
                        "content_pct": 1.7,
                        "tool_call_pct": 22.1
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 40.385,
                "latency": 2168.413,
                "stderr": 3.41,
                "cost_per_test": 0.838645,
                "token_totals": {
                    "input_tokens": 683339325,
                    "output_tokens": 23833353,
                    "reasoning_tokens": 0,
                    "cache_read_tokens": 604305152,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 47.8,
                    "avg_query_seconds": 45.3,
                    "median_query_seconds": 17.0,
                    "avg_ttft_seconds": 4.2,
                    "output_time_split": {
                        "reasoning_pct": 69.1,
                        "content_pct": 2.5,
                        "tool_call_pct": 28.4
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 40.385,
                "latency": 2265.105,
                "stderr": 3.41,
                "cost_per_test": 7.402471,
                "token_totals": {
                    "input_tokens": 743971258,
                    "output_tokens": 21289365,
                    "reasoning_tokens": 16694105,
                    "cache_read_tokens": 647774464,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 60.3,
                    "avg_query_seconds": 38.0,
                    "median_query_seconds": 6.6,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 39.423,
                "latency": 1491.339,
                "stderr": 3.397,
                "cost_per_test": 10.577143,
                "token_totals": {
                    "input_tokens": 290661092,
                    "output_tokens": 21448775,
                    "reasoning_tokens": 16238030,
                    "cache_read_tokens": 215071044,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 19.0,
                    "avg_query_seconds": 78.6,
                    "median_query_seconds": 31.7,
                    "avg_ttft_seconds": 9.4,
                    "output_time_split": {
                        "reasoning_pct": 25.5,
                        "content_pct": 0.6,
                        "tool_call_pct": 74.0
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 38.942,
                "latency": 324.318,
                "stderr": 3.389,
                "cost_per_test": 1.630498,
                "token_totals": {
                    "input_tokens": 581263116,
                    "output_tokens": 14420656,
                    "reasoning_tokens": 10782340,
                    "cache_read_tokens": 474745174,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 50.6,
                    "avg_query_seconds": 6.4,
                    "median_query_seconds": 1.8,
                    "avg_ttft_seconds": 1.3,
                    "output_time_split": {
                        "reasoning_pct": 98.4,
                        "content_pct": 0.0,
                        "tool_call_pct": 1.6
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 38.462,
                "latency": 511.955,
                "stderr": 3.381,
                "cost_per_test": 3.05609,
                "token_totals": {
                    "input_tokens": 280128756,
                    "output_tokens": 7275719,
                    "reasoning_tokens": 2846795,
                    "cache_read_tokens": 225999114,
                    "cache_write_tokens": 53655955
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 19.8,
                    "avg_query_seconds": 25.9,
                    "median_query_seconds": 13.9,
                    "avg_ttft_seconds": 8.5,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 38.462,
                "latency": 989.604,
                "stderr": 3.381,
                "cost_per_test": 2.302948,
                "token_totals": {
                    "input_tokens": 294037202,
                    "output_tokens": 10695335,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 223683365,
                    "cache_write_tokens": 70074508
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 24.6,
                    "avg_query_seconds": 40.0,
                    "median_query_seconds": 20.9,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 38.462,
                "latency": 1763.023,
                "stderr": 3.381,
                "cost_per_test": 1.559385,
                "token_totals": {
                    "input_tokens": 1351006836,
                    "output_tokens": 6986771,
                    "reasoning_tokens": 3248800,
                    "cache_read_tokens": 1267363808,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 31.6,
                    "avg_query_seconds": 55.8,
                    "median_query_seconds": 10.0,
                    "avg_ttft_seconds": 11.0,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 38.462,
                "latency": 2060.48,
                "stderr": 3.381,
                "cost_per_test": 3.384561,
                "token_totals": {
                    "input_tokens": 329856713,
                    "output_tokens": 20763888,
                    "reasoning_tokens": 15434407,
                    "cache_read_tokens": 260063643,
                    "cache_write_tokens": 68892844
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 21.7,
                    "avg_query_seconds": 95.1,
                    "median_query_seconds": 43.9,
                    "avg_ttft_seconds": 13.1,
                    "output_time_split": {
                        "reasoning_pct": 24.3,
                        "content_pct": 0.9,
                        "tool_call_pct": 74.9
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 37.981,
                "latency": 312.994,
                "stderr": 3.373,
                "cost_per_test": 0.377379,
                "token_totals": {
                    "input_tokens": 126240889,
                    "output_tokens": 5901711,
                    "reasoning_tokens": 3106623,
                    "cache_read_tokens": 94898720,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 16.0,
                    "avg_query_seconds": 19.6,
                    "median_query_seconds": 9.4,
                    "avg_ttft_seconds": 4.1,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 37.981,
                "latency": 919.689,
                "stderr": 3.373,
                "cost_per_test": 0.070424,
                "token_totals": {
                    "input_tokens": 365192541,
                    "output_tokens": 17884630,
                    "reasoning_tokens": 10104423,
                    "cache_read_tokens": 305459072,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 202,
                "turn_metrics": {
                    "tasks_measured": 205,
                    "avg_turns": 27.4,
                    "avg_query_seconds": 35.9,
                    "median_query_seconds": 19.2,
                    "avg_ttft_seconds": 3.5,
                    "output_time_split": {
                        "reasoning_pct": 0.7,
                        "content_pct": 0.0,
                        "tool_call_pct": 99.3
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 37.981,
                "latency": 953.767,
                "stderr": 3.373,
                "cost_per_test": 1.086814,
                "token_totals": {
                    "input_tokens": 128982108,
                    "output_tokens": 6582496,
                    "reasoning_tokens": 3272884,
                    "cache_read_tokens": 74664576,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 12.5,
                    "avg_query_seconds": 76.2,
                    "median_query_seconds": 26.8,
                    "avg_ttft_seconds": 16.0,
                    "output_time_split": null
                },
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 37.5,
                "latency": 6237.394,
                "stderr": 3.365,
                "cost_per_test": 7.231463,
                "token_totals": {
                    "input_tokens": 1110222674,
                    "output_tokens": 53232219,
                    "reasoning_tokens": 44000137,
                    "cache_read_tokens": 842326272,
                    "cache_write_tokens": 266448109
                },
                "token_tasks_measured": 171,
                "turn_metrics": {
                    "tasks_measured": 171,
                    "avg_turns": 44.1,
                    "avg_query_seconds": 141.6,
                    "median_query_seconds": 25.7,
                    "avg_ttft_seconds": 7.8,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 36.538,
                "latency": 2373.859,
                "stderr": 3.347,
                "cost_per_test": 0.845008,
                "token_totals": {
                    "input_tokens": 1755111665,
                    "output_tokens": 88330527,
                    "reasoning_tokens": 78699683,
                    "cache_read_tokens": 1562540626,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 53.5,
                    "avg_query_seconds": 44.3,
                    "median_query_seconds": 7.5,
                    "avg_ttft_seconds": 3.4,
                    "output_time_split": {
                        "reasoning_pct": 45.7,
                        "content_pct": 1.1,
                        "tool_call_pct": 53.2
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 36.058,
                "latency": 1600.721,
                "stderr": 3.337,
                "cost_per_test": 2.159719,
                "token_totals": {
                    "input_tokens": 600057696,
                    "output_tokens": 49730866,
                    "reasoning_tokens": 42100615,
                    "cache_read_tokens": 0,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 34.9,
                    "avg_query_seconds": 45.8,
                    "median_query_seconds": 17.0,
                    "avg_ttft_seconds": 3.1,
                    "output_time_split": {
                        "reasoning_pct": 78.0,
                        "content_pct": 1.5,
                        "tool_call_pct": 20.5
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 35.577,
                "latency": 1704.14,
                "stderr": 3.328,
                "cost_per_test": 1.09052,
                "token_totals": {
                    "input_tokens": 803000611,
                    "output_tokens": 18904714,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 595193856,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 30.3,
                    "avg_query_seconds": 56.2,
                    "median_query_seconds": 24.4,
                    "avg_ttft_seconds": 4.6,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 34.615,
                "latency": 288.96,
                "stderr": 3.307,
                "cost_per_test": 0.760397,
                "token_totals": {
                    "input_tokens": 216485075,
                    "output_tokens": 6173336,
                    "reasoning_tokens": 3838872,
                    "cache_read_tokens": 157677752,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 29.7,
                    "avg_query_seconds": 9.7,
                    "median_query_seconds": 2.8,
                    "avg_ttft_seconds": 2.3,
                    "output_time_split": {
                        "reasoning_pct": 98.0,
                        "content_pct": 0.0,
                        "tool_call_pct": 2.0
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 34.135,
                "latency": 2911.983,
                "stderr": 3.296,
                "cost_per_test": 0.444777,
                "token_totals": {
                    "input_tokens": 1230685744,
                    "output_tokens": 102073269,
                    "reasoning_tokens": 90583933,
                    "cache_read_tokens": 1135801046,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 54.4,
                    "avg_query_seconds": 53.5,
                    "median_query_seconds": 16.7,
                    "avg_ttft_seconds": 1.3,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 31.731,
                "latency": 5571.099,
                "stderr": 3.235,
                "cost_per_test": 2.723732,
                "token_totals": {
                    "input_tokens": 1494014814,
                    "output_tokens": 37508841,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 885782016,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 46.9,
                    "avg_query_seconds": 118.8,
                    "median_query_seconds": 45.4,
                    "avg_ttft_seconds": 10.6,
                    "output_time_split": {
                        "reasoning_pct": 88.6,
                        "content_pct": 11.2,
                        "tool_call_pct": 0.2
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 31.25,
                "latency": 1023.468,
                "stderr": 3.222,
                "cost_per_test": 0.889209,
                "token_totals": {
                    "input_tokens": 268943504,
                    "output_tokens": 12537934,
                    "reasoning_tokens": 9395692,
                    "cache_read_tokens": 210799936,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 23.3,
                    "avg_query_seconds": 44.9,
                    "median_query_seconds": 24.0,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 30.769,
                "latency": 277.753,
                "stderr": 3.208,
                "cost_per_test": 1.305281,
                "token_totals": {
                    "input_tokens": 556323486,
                    "output_tokens": 7365405,
                    "reasoning_tokens": 5568107,
                    "cache_read_tokens": 472254997,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 45.5,
                    "avg_query_seconds": 5.9,
                    "median_query_seconds": 2.6,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 30.288,
                "latency": 1266.744,
                "stderr": 3.194,
                "cost_per_test": 0.275172,
                "token_totals": {
                    "input_tokens": 310561183,
                    "output_tokens": 24287114,
                    "reasoning_tokens": 20969113,
                    "cache_read_tokens": 261666816,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 26.3,
                    "avg_query_seconds": 48.2,
                    "median_query_seconds": 18.4,
                    "avg_ttft_seconds": 3.9,
                    "output_time_split": {
                        "reasoning_pct": 75.4,
                        "content_pct": 2.6,
                        "tool_call_pct": 21.9
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 30.288,
                "latency": 2547.747,
                "stderr": 3.194,
                "cost_per_test": 0.440296,
                "token_totals": {
                    "input_tokens": 1143721770,
                    "output_tokens": 59694127,
                    "reasoning_tokens": 54603753,
                    "cache_read_tokens": 1042899991,
                    "cache_write_tokens": 99842719
                },
                "token_tasks_measured": 206,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 38.0,
                    "avg_query_seconds": 67.1,
                    "median_query_seconds": 16.9,
                    "avg_ttft_seconds": 7.2,
                    "output_time_split": {
                        "reasoning_pct": 66.4,
                        "content_pct": 0.0,
                        "tool_call_pct": 33.6
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 29.808,
                "latency": 813.838,
                "stderr": 3.179,
                "cost_per_test": 0.342955,
                "token_totals": {
                    "input_tokens": 273575187,
                    "output_tokens": 6440429,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 219245518,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 26.9,
                    "avg_query_seconds": 30.5,
                    "median_query_seconds": 12.3,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 28.846,
                "latency": 94.568,
                "stderr": 3.149,
                "cost_per_test": null,
                "token_totals": {
                    "input_tokens": 394030584,
                    "output_tokens": 5402310,
                    "reasoning_tokens": 3403372,
                    "cache_read_tokens": 359513792,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 36.4,
                    "avg_query_seconds": 2.6,
                    "median_query_seconds": 1.3,
                    "avg_ttft_seconds": 0.4,
                    "output_time_split": null
                },
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 28.846,
                "latency": 446.517,
                "stderr": 3.149,
                "cost_per_test": 0.074351,
                "token_totals": {
                    "input_tokens": 271885585,
                    "output_tokens": 22851545,
                    "reasoning_tokens": 20198714,
                    "cache_read_tokens": 219955154,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 26.3,
                    "avg_query_seconds": 16.9,
                    "median_query_seconds": 6.1,
                    "avg_ttft_seconds": 3.3,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 28.846,
                "latency": 1449.451,
                "stderr": 3.149,
                "cost_per_test": 4.897717,
                "token_totals": {
                    "input_tokens": 709313788,
                    "output_tokens": 26332564,
                    "reasoning_tokens": 22035035,
                    "cache_read_tokens": 634566483,
                    "cache_write_tokens": 73881332
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 35.7,
                    "avg_query_seconds": 40.6,
                    "median_query_seconds": 10.6,
                    "avg_ttft_seconds": 5.5,
                    "output_time_split": {
                        "reasoning_pct": 33.1,
                        "content_pct": 2.4,
                        "tool_call_pct": 64.5
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 28.365,
                "latency": 869.377,
                "stderr": 3.133,
                "cost_per_test": 0.494044,
                "token_totals": {
                    "input_tokens": 271552348,
                    "output_tokens": 8856526,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 246578560,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 37.8,
                    "avg_query_seconds": 23.0,
                    "median_query_seconds": 4.5,
                    "avg_ttft_seconds": 2.7,
                    "output_time_split": {
                        "reasoning_pct": 99.9,
                        "content_pct": 0.0,
                        "tool_call_pct": 0.1
                    }
                },
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 27.885,
                "latency": 325.824,
                "stderr": 3.117,
                "cost_per_test": 0.385737,
                "token_totals": {
                    "input_tokens": 185419655,
                    "output_tokens": 2414644,
                    "reasoning_tokens": 0,
                    "cache_read_tokens": 140363008,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 25.1,
                    "avg_query_seconds": 12.9,
                    "median_query_seconds": 6.2,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 25.962,
                "latency": 4202.728,
                "stderr": 3.047,
                "cost_per_test": 3.220614,
                "token_totals": {
                    "input_tokens": 1392458117,
                    "output_tokens": 69708508,
                    "reasoning_tokens": 62148334,
                    "cache_read_tokens": 1260829967,
                    "cache_write_tokens": 130975483
                },
                "token_tasks_measured": 202,
                "turn_metrics": {
                    "tasks_measured": 202,
                    "avg_turns": 35.9,
                    "avg_query_seconds": 117.1,
                    "median_query_seconds": 21.4,
                    "avg_ttft_seconds": 8.0,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 25.481,
                "latency": 774.823,
                "stderr": 3.029,
                "cost_per_test": 0.141548,
                "token_totals": {
                    "input_tokens": 209190661,
                    "output_tokens": 10262319,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 190714496,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 34.6,
                    "avg_query_seconds": 22.4,
                    "median_query_seconds": 5.1,
                    "avg_ttft_seconds": 1.9,
                    "output_time_split": {
                        "reasoning_pct": 100.0,
                        "content_pct": 0.0,
                        "tool_call_pct": 0.0
                    }
                },
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 25.481,
                "latency": 808.802,
                "stderr": 3.029,
                "cost_per_test": 1.400701,
                "token_totals": {
                    "input_tokens": 158509885,
                    "output_tokens": 6696005,
                    "reasoning_tokens": 4196744,
                    "cache_read_tokens": 68025984,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 19.7,
                    "avg_query_seconds": 41.6,
                    "median_query_seconds": 10.1,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 25.0,
                "latency": 168.482,
                "stderr": 3.01,
                "cost_per_test": 0.466741,
                "token_totals": {
                    "input_tokens": 116699947,
                    "output_tokens": 4157671,
                    "reasoning_tokens": 2875086,
                    "cache_read_tokens": 80852090,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 20.1,
                    "avg_query_seconds": 8.4,
                    "median_query_seconds": 3.2,
                    "avg_ttft_seconds": 1.5,
                    "output_time_split": {
                        "reasoning_pct": 99.4,
                        "content_pct": 0.0,
                        "tool_call_pct": 0.6
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 23.077,
                "latency": 1024.619,
                "stderr": 2.928,
                "cost_per_test": 0.682218,
                "token_totals": {
                    "input_tokens": 341401669,
                    "output_tokens": 11003300,
                    "reasoning_tokens": 8033786,
                    "cache_read_tokens": 300570368,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 32.8,
                    "avg_query_seconds": 34.4,
                    "median_query_seconds": 8.3,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 20.673,
                "latency": 242.05,
                "stderr": 2.815,
                "cost_per_test": 1.100145,
                "token_totals": {
                    "input_tokens": 354575717,
                    "output_tokens": 2983139,
                    "reasoning_tokens": 2206660,
                    "cache_read_tokens": 321431187,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 22.8,
                    "avg_query_seconds": 10.5,
                    "median_query_seconds": 6.7,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 18.269,
                "latency": 187.7,
                "stderr": 2.686,
                "cost_per_test": 0.135395,
                "token_totals": {
                    "input_tokens": 80643637,
                    "output_tokens": 3487795,
                    "reasoning_tokens": 2539699,
                    "cache_read_tokens": 50273506,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 14.7,
                    "avg_query_seconds": 12.7,
                    "median_query_seconds": 6.9,
                    "avg_ttft_seconds": 2.1,
                    "output_time_split": {
                        "reasoning_pct": 87.9,
                        "content_pct": 11.4,
                        "tool_call_pct": 0.7
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 16.346,
                "latency": 831.995,
                "stderr": 2.57,
                "cost_per_test": 0.295949,
                "token_totals": {
                    "input_tokens": 182291959,
                    "output_tokens": 7407756,
                    "reasoning_tokens": 4986270,
                    "cache_read_tokens": 118507648,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 206,
                "turn_metrics": {
                    "tasks_measured": 206,
                    "avg_turns": 22.5,
                    "avg_query_seconds": 36.9,
                    "median_query_seconds": 12.5,
                    "avg_ttft_seconds": 3.5,
                    "output_time_split": {
                        "reasoning_pct": 49.4,
                        "content_pct": 3.3,
                        "tool_call_pct": 47.3
                    }
                },
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 15.865,
                "latency": 416.437,
                "stderr": 2.539,
                "cost_per_test": 0.082648,
                "token_totals": {
                    "input_tokens": 179797926,
                    "output_tokens": 3117976,
                    "reasoning_tokens": 0,
                    "cache_read_tokens": 147737472,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 24.1,
                    "avg_query_seconds": 17.3,
                    "median_query_seconds": 8.4,
                    "avg_ttft_seconds": 3.8,
                    "output_time_split": {
                        "reasoning_pct": 44.7,
                        "content_pct": 1.5,
                        "tool_call_pct": 53.8
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 15.865,
                "latency": 1332.514,
                "stderr": 2.539,
                "cost_per_test": 0.23035,
                "token_totals": {
                    "input_tokens": 207645988,
                    "output_tokens": 4488005,
                    "reasoning_tokens": 2618301,
                    "cache_read_tokens": 180277739,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 27.5,
                    "avg_query_seconds": 48.5,
                    "median_query_seconds": 36.8,
                    "avg_ttft_seconds": 23.5,
                    "output_time_split": {
                        "reasoning_pct": 51.3,
                        "content_pct": 12.7,
                        "tool_call_pct": 36.1
                    }
                },
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 15.865,
                "latency": 1853.133,
                "stderr": 2.539,
                "cost_per_test": 1.520305,
                "token_totals": {
                    "input_tokens": 1570147541,
                    "output_tokens": 3643156,
                    "reasoning_tokens": 0,
                    "cache_read_tokens": 1492103312,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 92.8,
                    "avg_query_seconds": 20.5,
                    "median_query_seconds": 5.8,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 15.385,
                "latency": 84.583,
                "stderr": 2.508,
                "cost_per_test": 0.203143,
                "token_totals": {
                    "input_tokens": 65647893,
                    "output_tokens": 3371863,
                    "reasoning_tokens": 2425742,
                    "cache_read_tokens": 48015040,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 14.4,
                    "avg_query_seconds": 6.4,
                    "median_query_seconds": 3.3,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 15.385,
                "latency": 515.464,
                "stderr": 2.508,
                "cost_per_test": 0.028227,
                "token_totals": {
                    "input_tokens": 221182940,
                    "output_tokens": 7577859,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 182576832,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 26.5,
                    "avg_query_seconds": 19.5,
                    "median_query_seconds": 8.9,
                    "avg_ttft_seconds": 5.6,
                    "output_time_split": {
                        "reasoning_pct": 59.5,
                        "content_pct": 2.9,
                        "tool_call_pct": 37.6
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 15.385,
                "latency": 611.366,
                "stderr": 2.508,
                "cost_per_test": null,
                "token_totals": {
                    "input_tokens": 181276135,
                    "output_tokens": 3143670,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 49852416,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 205,
                "turn_metrics": {
                    "tasks_measured": 205,
                    "avg_turns": 29.6,
                    "avg_query_seconds": 20.6,
                    "median_query_seconds": 9.4,
                    "avg_ttft_seconds": 7.9,
                    "output_time_split": {
                        "reasoning_pct": 52.8,
                        "content_pct": 14.4,
                        "tool_call_pct": 32.9
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 14.904,
                "latency": 658.954,
                "stderr": 2.475,
                "cost_per_test": 0.697997,
                "token_totals": {
                    "input_tokens": 228091457,
                    "output_tokens": 5468376,
                    "reasoning_tokens": 3418357,
                    "cache_read_tokens": 0,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 26.6,
                    "avg_query_seconds": 24.8,
                    "median_query_seconds": 9.9,
                    "avg_ttft_seconds": 4.0,
                    "output_time_split": {
                        "reasoning_pct": 42.7,
                        "content_pct": 1.2,
                        "tool_call_pct": 56.1
                    }
                },
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 13.942,
                "latency": 59.197,
                "stderr": 2.408,
                "cost_per_test": 0.124723,
                "token_totals": {
                    "input_tokens": 121402055,
                    "output_tokens": 1885651,
                    "reasoning_tokens": 1270438,
                    "cache_read_tokens": 56268106,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 19.0,
                    "avg_query_seconds": 3.1,
                    "median_query_seconds": 1.9,
                    "avg_ttft_seconds": 1.1,
                    "output_time_split": {
                        "reasoning_pct": 90.3,
                        "content_pct": 8.6,
                        "tool_call_pct": 1.1
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 13.942,
                "latency": 244.487,
                "stderr": 2.408,
                "cost_per_test": 0.261499,
                "token_totals": {
                    "input_tokens": 45228797,
                    "output_tokens": 1394502,
                    "reasoning_tokens": 691251,
                    "cache_read_tokens": 30303746,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 12.5,
                    "avg_query_seconds": 19.6,
                    "median_query_seconds": 4.0,
                    "avg_ttft_seconds": 5.9,
                    "output_time_split": {
                        "reasoning_pct": 58.2,
                        "content_pct": 2.7,
                        "tool_call_pct": 39.0
                    }
                },
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 12.5,
                "latency": 2256.339,
                "stderr": 2.299,
                "cost_per_test": 1.864673,
                "token_totals": {
                    "input_tokens": 742970115,
                    "output_tokens": 55568745,
                    "reasoning_tokens": 53294787,
                    "cache_read_tokens": 622977536,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 45.0,
                    "avg_query_seconds": 50.1,
                    "median_query_seconds": 8.7,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 10.577,
                "latency": 243.38,
                "stderr": 2.138,
                "cost_per_test": 0.26891,
                "token_totals": {
                    "input_tokens": 131779295,
                    "output_tokens": 2744297,
                    "reasoning_tokens": 1091315,
                    "cache_read_tokens": 106353006,
                    "cache_write_tokens": 24601089
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 17.2,
                    "avg_query_seconds": 14.2,
                    "median_query_seconds": 6.8,
                    "avg_ttft_seconds": 1.2,
                    "output_time_split": {
                        "reasoning_pct": 49.2,
                        "content_pct": 16.8,
                        "tool_call_pct": 34.0
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 10.577,
                "latency": 420.647,
                "stderr": 2.138,
                "cost_per_test": 0.219167,
                "token_totals": {
                    "input_tokens": 143277305,
                    "output_tokens": 5052868,
                    "reasoning_tokens": 3440031,
                    "cache_read_tokens": 14839186,
                    "cache_write_tokens": 7287836
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 26.4,
                    "avg_query_seconds": 15.9,
                    "median_query_seconds": 5.0,
                    "avg_ttft_seconds": 2.1,
                    "output_time_split": {
                        "reasoning_pct": 55.4,
                        "content_pct": 8.9,
                        "tool_call_pct": 35.8
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 10.096,
                "latency": 82.971,
                "stderr": 2.094,
                "cost_per_test": 0.063761,
                "token_totals": {
                    "input_tokens": 92421794,
                    "output_tokens": 5513825,
                    "reasoning_tokens": 3920900,
                    "cache_read_tokens": 49868032,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 16.2,
                    "avg_query_seconds": 5.1,
                    "median_query_seconds": 2.7,
                    "avg_ttft_seconds": 0.5,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 9.135,
                "latency": 391.469,
                "stderr": 2.002,
                "cost_per_test": 0.033318,
                "token_totals": {
                    "input_tokens": 251871878,
                    "output_tokens": 3303008,
                    "reasoning_tokens": 1325334,
                    "cache_read_tokens": 213241152,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 35.4,
                    "avg_query_seconds": 11.1,
                    "median_query_seconds": 6.6,
                    "avg_ttft_seconds": 3.9,
                    "output_time_split": {
                        "reasoning_pct": 52.6,
                        "content_pct": 5.3,
                        "tool_call_pct": 42.1
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 9.135,
                "latency": 613.248,
                "stderr": 2.002,
                "cost_per_test": 1.308079,
                "token_totals": {
                    "input_tokens": 148782567,
                    "output_tokens": 6520890,
                    "reasoning_tokens": null,
                    "cache_read_tokens": null,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 25.9,
                    "avg_query_seconds": 23.7,
                    "median_query_seconds": 9.4,
                    "avg_ttft_seconds": 6.0,
                    "output_time_split": {
                        "reasoning_pct": 85.8,
                        "content_pct": 12.5,
                        "tool_call_pct": 1.7
                    }
                },
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 6.25,
                "latency": 526.838,
                "stderr": 1.682,
                "cost_per_test": 0.141253,
                "token_totals": {
                    "input_tokens": 386062284,
                    "output_tokens": 6536232,
                    "reasoning_tokens": 4445328,
                    "cache_read_tokens": 311122688,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 40.6,
                    "avg_query_seconds": 13.0,
                    "median_query_seconds": 6.7,
                    "avg_ttft_seconds": 3.5,
                    "output_time_split": {
                        "reasoning_pct": 71.7,
                        "content_pct": 0.0,
                        "tool_call_pct": 28.3
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 4.327,
                "latency": 47.609,
                "stderr": 1.414,
                "cost_per_test": 0.058112,
                "token_totals": {
                    "input_tokens": 77109229,
                    "output_tokens": 1908210,
                    "reasoning_tokens": 1539184,
                    "cache_read_tokens": 25000429,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 15.4,
                    "avg_query_seconds": 3.1,
                    "median_query_seconds": 2.0,
                    "avg_ttft_seconds": 1.6,
                    "output_time_split": null
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 3.365,
                "latency": 20.403,
                "stderr": 1.253,
                "cost_per_test": 0.013607,
                "token_totals": {
                    "input_tokens": 8137148,
                    "output_tokens": 895676,
                    "reasoning_tokens": 458069,
                    "cache_read_tokens": 2433959,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 4.3,
                    "avg_query_seconds": 4.8,
                    "median_query_seconds": 4.2,
                    "avg_ttft_seconds": 1.4,
                    "output_time_split": {
                        "reasoning_pct": 65.6,
                        "content_pct": 33.7,
                        "tool_call_pct": 0.7
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 2.885,
                "latency": 285.48,
                "stderr": 1.163,
                "cost_per_test": 0.004365,
                "token_totals": {
                    "input_tokens": 24564654,
                    "output_tokens": 1677472,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 16501798,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 207,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 9.3,
                    "avg_query_seconds": 30.4,
                    "median_query_seconds": 2.2,
                    "avg_ttft_seconds": 14.5,
                    "output_time_split": {
                        "reasoning_pct": 68.7,
                        "content_pct": 28.4,
                        "tool_call_pct": 2.9
                    }
                },
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 2.404,
                "latency": 466.26,
                "stderr": 1.065,
                "cost_per_test": 0.219839,
                "token_totals": {
                    "input_tokens": 388327686,
                    "output_tokens": 3513231,
                    "reasoning_tokens": 1795689,
                    "cache_read_tokens": 333442324,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 44.2,
                    "avg_query_seconds": 10.5,
                    "median_query_seconds": 4.7,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.962,
                "latency": 352.036,
                "stderr": 0.678,
                "cost_per_test": 0.107892,
                "token_totals": {
                    "input_tokens": 363797527,
                    "output_tokens": 2340571,
                    "reasoning_tokens": 1061598,
                    "cache_read_tokens": 288125104,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 208,
                "turn_metrics": {
                    "tasks_measured": 208,
                    "avg_turns": 41.7,
                    "avg_query_seconds": 8.4,
                    "median_query_seconds": 3.2,
                    "avg_ttft_seconds": null,
                    "output_time_split": null
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 211.833,
                "stderr": 0.0,
                "cost_per_test": 0.036023,
                "token_totals": {
                    "input_tokens": 190331294,
                    "output_tokens": 4335412,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 130131626,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 206,
                "turn_metrics": {
                    "tasks_measured": 207,
                    "avg_turns": 21.3,
                    "avg_query_seconds": 9.9,
                    "median_query_seconds": 5.3,
                    "avg_ttft_seconds": 2.4,
                    "output_time_split": {
                        "reasoning_pct": 66.4,
                        "content_pct": 2.6,
                        "tool_call_pct": 31.0
                    }
                },
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            }
        },
        "administrative_regulatory": {
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 69.697,
                "latency": 4690.9,
                "stderr": 5.7,
                "cost_per_test": 11.667322,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 68.182,
                "latency": 1486.33,
                "stderr": 5.777,
                "cost_per_test": 4.807402,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 68.182,
                "latency": 3106.859,
                "stderr": 5.777,
                "cost_per_test": 20.305794,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 65.152,
                "latency": 1532.62,
                "stderr": 5.91,
                "cost_per_test": 6.583578,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 65.152,
                "latency": 2535.332,
                "stderr": 5.91,
                "cost_per_test": 6.803146,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 65.152,
                "latency": 5979.926,
                "stderr": 5.91,
                "cost_per_test": 20.862676,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 63.636,
                "latency": 749.583,
                "stderr": 5.967,
                "cost_per_test": 3.444134,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 63.636,
                "latency": 889.246,
                "stderr": 5.967,
                "cost_per_test": 3.922674,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 63.636,
                "latency": 1691.752,
                "stderr": 5.967,
                "cost_per_test": 3.18787,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 63.636,
                "latency": 2038.532,
                "stderr": 5.967,
                "cost_per_test": 3.174554,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 62.121,
                "latency": 1372.59,
                "stderr": 6.017,
                "cost_per_test": 1.517644,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 62.121,
                "latency": 2956.379,
                "stderr": 6.017,
                "cost_per_test": 2.133253,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 60.606,
                "latency": 306.876,
                "stderr": 6.061,
                "cost_per_test": 0.540035,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 60.606,
                "latency": 631.256,
                "stderr": 6.061,
                "cost_per_test": 0.243078,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 60.606,
                "latency": 4138.255,
                "stderr": 6.061,
                "cost_per_test": 19.689902,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 59.091,
                "latency": 1609.837,
                "stderr": 6.098,
                "cost_per_test": 0.17954,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 59.091,
                "latency": 1931.905,
                "stderr": 6.098,
                "cost_per_test": 0.652538,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 59.091,
                "latency": 3192.985,
                "stderr": 6.098,
                "cost_per_test": 2.201925,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 57.576,
                "latency": 1179.378,
                "stderr": 6.13,
                "cost_per_test": 9.381311,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 57.576,
                "latency": 2848.785,
                "stderr": 6.13,
                "cost_per_test": 0.198792,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 57.576,
                "latency": 3607.895,
                "stderr": 6.13,
                "cost_per_test": 6.866314,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 56.061,
                "latency": 5673.421,
                "stderr": 6.156,
                "cost_per_test": 6.338985,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 54.545,
                "latency": 1016.365,
                "stderr": 6.176,
                "cost_per_test": 4.175009,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 54.545,
                "latency": 1285.722,
                "stderr": 6.176,
                "cost_per_test": 2.507811,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 54.545,
                "latency": 3713.178,
                "stderr": 6.176,
                "cost_per_test": 0.834838,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 53.03,
                "latency": 656.731,
                "stderr": 6.19,
                "cost_per_test": 0.246492,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 53.03,
                "latency": 1103.075,
                "stderr": 6.19,
                "cost_per_test": 8.442858,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 53.03,
                "latency": 1439.965,
                "stderr": 6.19,
                "cost_per_test": 2.756725,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 53.03,
                "latency": 2093.383,
                "stderr": 6.19,
                "cost_per_test": 7.953883,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 51.515,
                "latency": 481.922,
                "stderr": 6.199,
                "cost_per_test": 0.624599,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 51.515,
                "latency": 1001.46,
                "stderr": 6.199,
                "cost_per_test": 1.36196,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 51.515,
                "latency": 5301.951,
                "stderr": 6.199,
                "cost_per_test": 2.211937,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 50.0,
                "latency": 1671.25,
                "stderr": 6.202,
                "cost_per_test": 1.423111,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 50.0,
                "latency": 2424.284,
                "stderr": 6.202,
                "cost_per_test": 1.073701,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 50.0,
                "latency": 3941.13,
                "stderr": 6.202,
                "cost_per_test": 1.441703,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 50.0,
                "latency": 7602.624,
                "stderr": 6.202,
                "cost_per_test": 5.772668,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 48.485,
                "latency": 302.672,
                "stderr": 6.199,
                "cost_per_test": 0.506511,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 48.485,
                "latency": 2171.348,
                "stderr": 6.199,
                "cost_per_test": 0.862352,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 48.485,
                "latency": 2245.491,
                "stderr": 6.199,
                "cost_per_test": 0.755621,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 46.97,
                "latency": 296.369,
                "stderr": 6.19,
                "cost_per_test": 1.661721,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 46.97,
                "latency": 715.742,
                "stderr": 6.19,
                "cost_per_test": 0.066967,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 46.97,
                "latency": 1523.131,
                "stderr": 6.19,
                "cost_per_test": 0.921373,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 46.97,
                "latency": 2358.521,
                "stderr": 6.19,
                "cost_per_test": 0.049642,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 45.455,
                "latency": 425.156,
                "stderr": 6.176,
                "cost_per_test": 3.06182,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 45.455,
                "latency": 1643.295,
                "stderr": 6.176,
                "cost_per_test": 2.2119,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 45.455,
                "latency": 3381.142,
                "stderr": 6.176,
                "cost_per_test": 2.810214,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 43.939,
                "latency": 285.017,
                "stderr": 6.156,
                "cost_per_test": 1.356228,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 43.939,
                "latency": 636.543,
                "stderr": 6.156,
                "cost_per_test": 2.731524,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 43.939,
                "latency": 901.446,
                "stderr": 6.156,
                "cost_per_test": 2.410326,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 42.424,
                "latency": 94.03,
                "stderr": 6.13,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 42.424,
                "latency": 244.664,
                "stderr": 6.13,
                "cost_per_test": 0.704178,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 40.909,
                "latency": 2966.067,
                "stderr": 6.098,
                "cost_per_test": 0.472211,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 39.394,
                "latency": 836.031,
                "stderr": 6.061,
                "cost_per_test": 0.808378,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 39.394,
                "latency": 1112.655,
                "stderr": 6.061,
                "cost_per_test": 3.887103,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 37.879,
                "latency": 282.392,
                "stderr": 6.017,
                "cost_per_test": 0.382532,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 37.879,
                "latency": 2341.023,
                "stderr": 6.017,
                "cost_per_test": 0.409481,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 36.364,
                "latency": 755.036,
                "stderr": 5.967,
                "cost_per_test": 0.352926,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 36.364,
                "latency": 789.443,
                "stderr": 5.967,
                "cost_per_test": 1.359392,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 36.364,
                "latency": 1185.766,
                "stderr": 5.967,
                "cost_per_test": 0.295871,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 34.848,
                "latency": 434.539,
                "stderr": 5.91,
                "cost_per_test": 0.075638,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 34.848,
                "latency": 5582.268,
                "stderr": 5.91,
                "cost_per_test": 2.616141,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 31.818,
                "latency": 139.396,
                "stderr": 5.777,
                "cost_per_test": 0.409841,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 31.818,
                "latency": 307.232,
                "stderr": 5.777,
                "cost_per_test": 0.323431,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 31.818,
                "latency": 877.92,
                "stderr": 5.777,
                "cost_per_test": 0.44143,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 28.788,
                "latency": 842.128,
                "stderr": 5.616,
                "cost_per_test": 0.134944,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 28.788,
                "latency": 903.624,
                "stderr": 5.616,
                "cost_per_test": 0.671217,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 27.273,
                "latency": 166.672,
                "stderr": 5.524,
                "cost_per_test": 0.129186,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 25.758,
                "latency": 72.945,
                "stderr": 5.424,
                "cost_per_test": 0.19233,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 24.242,
                "latency": 219.629,
                "stderr": 5.316,
                "cost_per_test": 1.225685,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 19.697,
                "latency": 50.778,
                "stderr": 4.933,
                "cost_per_test": 0.123387,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 19.697,
                "latency": 625.818,
                "stderr": 4.933,
                "cost_per_test": 0.670062,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 18.182,
                "latency": 381.12,
                "stderr": 4.784,
                "cost_per_test": 0.087857,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 18.182,
                "latency": 494.675,
                "stderr": 4.784,
                "cost_per_test": 0.027887,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 18.182,
                "latency": 1689.904,
                "stderr": 4.784,
                "cost_per_test": 1.488125,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 16.667,
                "latency": 382.835,
                "stderr": 4.623,
                "cost_per_test": 0.200329,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 16.667,
                "latency": 476.589,
                "stderr": 4.623,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 16.667,
                "latency": 1153.879,
                "stderr": 4.623,
                "cost_per_test": 0.195218,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 15.152,
                "latency": 805.914,
                "stderr": 4.447,
                "cost_per_test": 0.359357,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 15.152,
                "latency": 1961.008,
                "stderr": 4.447,
                "cost_per_test": 1.456649,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 10.606,
                "latency": 199.066,
                "stderr": 3.819,
                "cost_per_test": 0.274657,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 10.606,
                "latency": 343.766,
                "stderr": 3.819,
                "cost_per_test": 0.038807,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 10.606,
                "latency": 402.197,
                "stderr": 3.819,
                "cost_per_test": 0.125736,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 9.091,
                "latency": 226.406,
                "stderr": 3.566,
                "cost_per_test": 0.379263,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 7.576,
                "latency": 575.292,
                "stderr": 3.282,
                "cost_per_test": 1.130658,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 6.061,
                "latency": 69.443,
                "stderr": 2.96,
                "cost_per_test": 0.05243,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 4.545,
                "latency": 41.907,
                "stderr": 2.584,
                "cost_per_test": 0.046871,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 3.03,
                "latency": 18.892,
                "stderr": 2.126,
                "cost_per_test": 0.013517,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 3.03,
                "latency": 383.823,
                "stderr": 2.126,
                "cost_per_test": 0.163947,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 1.515,
                "latency": 332.137,
                "stderr": 1.515,
                "cost_per_test": 0.087793,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 1.515,
                "latency": 409.746,
                "stderr": 1.515,
                "cost_per_test": 0.004424,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 215.774,
                "stderr": 0.0,
                "cost_per_test": 0.032184,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            }
        },
        "business_commercial": {
            "anthropic/claude-opus-5": {
                "accuracy": 62.791,
                "latency": 2163.785,
                "stderr": 7.458,
                "cost_per_test": 7.357666,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 62.791,
                "latency": 3855.165,
                "stderr": 7.458,
                "cost_per_test": 27.230592,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 58.14,
                "latency": 353.92,
                "stderr": 7.612,
                "cost_per_test": 0.626202,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 58.14,
                "latency": 1748.44,
                "stderr": 7.612,
                "cost_per_test": 3.900906,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 53.488,
                "latency": 4581.547,
                "stderr": 7.696,
                "cost_per_test": 19.023356,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 51.163,
                "latency": 319.151,
                "stderr": 7.713,
                "cost_per_test": 0.363621,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 51.163,
                "latency": 1482.098,
                "stderr": 7.713,
                "cost_per_test": 9.515091,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 48.837,
                "latency": 904.491,
                "stderr": 7.713,
                "cost_per_test": 3.808111,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 48.837,
                "latency": 1612.05,
                "stderr": 7.713,
                "cost_per_test": 1.54459,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 48.837,
                "latency": 7461.121,
                "stderr": 7.713,
                "cost_per_test": 26.294195,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 46.512,
                "latency": 755.219,
                "stderr": 7.696,
                "cost_per_test": 0.250157,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 46.512,
                "latency": 2252.454,
                "stderr": 7.696,
                "cost_per_test": 7.818685,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 46.512,
                "latency": 3261.478,
                "stderr": 7.696,
                "cost_per_test": 2.29039,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 46.512,
                "latency": 3764.53,
                "stderr": 7.696,
                "cost_per_test": 10.672138,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 46.512,
                "latency": 3920.461,
                "stderr": 7.696,
                "cost_per_test": 2.547702,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 46.512,
                "latency": 8276.057,
                "stderr": 7.696,
                "cost_per_test": 5.198048,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 44.186,
                "latency": 328.596,
                "stderr": 7.663,
                "cost_per_test": 0.506176,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 44.186,
                "latency": 1380.854,
                "stderr": 7.663,
                "cost_per_test": 4.555398,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 44.186,
                "latency": 1783.688,
                "stderr": 7.663,
                "cost_per_test": 0.170401,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 44.186,
                "latency": 2247.75,
                "stderr": 7.663,
                "cost_per_test": 0.705293,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 44.186,
                "latency": 3410.204,
                "stderr": 7.663,
                "cost_per_test": 0.266594,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 44.186,
                "latency": 4069.494,
                "stderr": 7.663,
                "cost_per_test": 0.863275,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 44.186,
                "latency": 5681.673,
                "stderr": 7.663,
                "cost_per_test": 1.820007,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 41.86,
                "latency": 315.567,
                "stderr": 7.612,
                "cost_per_test": 1.733887,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 41.86,
                "latency": 777.706,
                "stderr": 7.612,
                "cost_per_test": 2.888428,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 41.86,
                "latency": 1723.973,
                "stderr": 7.612,
                "cost_per_test": 2.311817,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 39.535,
                "latency": 694.588,
                "stderr": 7.544,
                "cost_per_test": 0.226487,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 39.535,
                "latency": 790.908,
                "stderr": 7.544,
                "cost_per_test": 3.239177,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 39.535,
                "latency": 1523.334,
                "stderr": 7.544,
                "cost_per_test": 1.137443,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 39.535,
                "latency": 2344.543,
                "stderr": 7.544,
                "cost_per_test": 0.811263,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 39.535,
                "latency": 2357.851,
                "stderr": 7.544,
                "cost_per_test": 3.455558,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 39.535,
                "latency": 5721.537,
                "stderr": 7.544,
                "cost_per_test": 2.262179,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 39.535,
                "latency": 6273.67,
                "stderr": 7.544,
                "cost_per_test": 15.137463,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 37.209,
                "latency": 298.916,
                "stderr": 7.458,
                "cost_per_test": 0.773586,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 37.209,
                "latency": 909.412,
                "stderr": 7.458,
                "cost_per_test": 0.066442,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 37.209,
                "latency": 2266.219,
                "stderr": 7.458,
                "cost_per_test": 6.684872,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 37.209,
                "latency": 2576.157,
                "stderr": 7.458,
                "cost_per_test": 0.05022,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 37.209,
                "latency": 2633.008,
                "stderr": 7.458,
                "cost_per_test": 0.372136,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 37.209,
                "latency": 2895.709,
                "stderr": 7.458,
                "cost_per_test": 4.184362,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 37.209,
                "latency": 4084.965,
                "stderr": 7.458,
                "cost_per_test": 7.302701,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 34.884,
                "latency": 709.804,
                "stderr": 7.354,
                "cost_per_test": 0.132473,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 34.884,
                "latency": 1023.821,
                "stderr": 7.354,
                "cost_per_test": 2.037941,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 34.884,
                "latency": 2499.931,
                "stderr": 7.354,
                "cost_per_test": 1.025341,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 32.558,
                "latency": 581.128,
                "stderr": 7.231,
                "cost_per_test": 0.642459,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 32.558,
                "latency": 626.261,
                "stderr": 7.231,
                "cost_per_test": 3.528557,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 32.558,
                "latency": 1735.645,
                "stderr": 7.231,
                "cost_per_test": 10.624987,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 32.558,
                "latency": 1790.227,
                "stderr": 7.231,
                "cost_per_test": 3.031928,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 32.558,
                "latency": 2334.81,
                "stderr": 7.231,
                "cost_per_test": 0.78052,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 30.233,
                "latency": 913.577,
                "stderr": 7.087,
                "cost_per_test": 0.893714,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 30.233,
                "latency": 1961.481,
                "stderr": 7.087,
                "cost_per_test": 1.153979,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 27.907,
                "latency": 317.652,
                "stderr": 6.921,
                "cost_per_test": 1.232191,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 27.907,
                "latency": 1106.364,
                "stderr": 6.921,
                "cost_per_test": 0.920113,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 27.907,
                "latency": 1674.291,
                "stderr": 6.921,
                "cost_per_test": 5.299121,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 27.907,
                "latency": 6424.655,
                "stderr": 6.921,
                "cost_per_test": 7.199044,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 25.581,
                "latency": 108.067,
                "stderr": 6.733,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 25.581,
                "latency": 498.212,
                "stderr": 6.733,
                "cost_per_test": 0.086143,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 25.581,
                "latency": 792.476,
                "stderr": 6.733,
                "cost_per_test": 0.529635,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 25.581,
                "latency": 867.939,
                "stderr": 6.733,
                "cost_per_test": 0.34538,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 25.581,
                "latency": 5387.125,
                "stderr": 6.733,
                "cost_per_test": 1.920252,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 23.256,
                "latency": 173.23,
                "stderr": 6.519,
                "cost_per_test": 0.487431,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 23.256,
                "latency": 216.623,
                "stderr": 6.519,
                "cost_per_test": 0.45295,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 23.256,
                "latency": 336.248,
                "stderr": 6.519,
                "cost_per_test": 0.399162,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 23.256,
                "latency": 917.178,
                "stderr": 6.519,
                "cost_per_test": 0.622362,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 23.256,
                "latency": 1087.928,
                "stderr": 6.519,
                "cost_per_test": 0.215579,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 23.256,
                "latency": 2678.679,
                "stderr": 6.519,
                "cost_per_test": 0.455548,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 20.93,
                "latency": 913.289,
                "stderr": 6.277,
                "cost_per_test": 1.493846,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 20.93,
                "latency": 1873.65,
                "stderr": 6.277,
                "cost_per_test": 1.432344,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 18.605,
                "latency": 200.727,
                "stderr": 6.005,
                "cost_per_test": 0.135638,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 16.279,
                "latency": 241.456,
                "stderr": 5.696,
                "cost_per_test": 0.168225,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 16.279,
                "latency": 663.553,
                "stderr": 5.696,
                "cost_per_test": 0.73626,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 16.279,
                "latency": 5207.222,
                "stderr": 5.696,
                "cost_per_test": 3.545917,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 13.953,
                "latency": 73.592,
                "stderr": 5.347,
                "cost_per_test": 0.170453,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 13.953,
                "latency": 343.417,
                "stderr": 5.347,
                "cost_per_test": 0.211419,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 13.953,
                "latency": 489.906,
                "stderr": 5.347,
                "cost_per_test": 0.032371,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 13.953,
                "latency": 787.349,
                "stderr": 5.347,
                "cost_per_test": 0.216443,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 13.953,
                "latency": 1493.936,
                "stderr": 5.347,
                "cost_per_test": 0.245468,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 11.628,
                "latency": 44.657,
                "stderr": 4.946,
                "cost_per_test": 0.057037,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 11.628,
                "latency": 108.913,
                "stderr": 4.946,
                "cost_per_test": 0.08423,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 11.628,
                "latency": 452.214,
                "stderr": 4.946,
                "cost_per_test": 0.075902,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 11.628,
                "latency": 620.168,
                "stderr": 4.946,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 11.628,
                "latency": 2506.739,
                "stderr": 4.946,
                "cost_per_test": 2.128528,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 9.302,
                "latency": 279.601,
                "stderr": 4.482,
                "cost_per_test": 0.268062,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 9.302,
                "latency": 405.023,
                "stderr": 4.482,
                "cost_per_test": 0.029283,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 9.302,
                "latency": 637.024,
                "stderr": 4.482,
                "cost_per_test": 1.423124,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 6.977,
                "latency": 62.398,
                "stderr": 3.931,
                "cost_per_test": 0.120286,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 6.977,
                "latency": 568.266,
                "stderr": 3.931,
                "cost_per_test": 0.150807,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 4.651,
                "latency": 124.79,
                "stderr": 3.249,
                "cost_per_test": 0.004634,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 4.651,
                "latency": 402.376,
                "stderr": 3.249,
                "cost_per_test": 0.204493,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 2.326,
                "latency": 23.436,
                "stderr": 2.326,
                "cost_per_test": 0.014473,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 261.305,
                "stderr": 0.0,
                "cost_per_test": 0.042315,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.0,
                "latency": 399.428,
                "stderr": 0.0,
                "cost_per_test": 0.12794,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            }
        },
        "civil_litigation": {
            "meta/muse_spark_1_3_max": {
                "accuracy": 62.963,
                "latency": 305.909,
                "stderr": 9.471,
                "cost_per_test": 0.581163,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 62.963,
                "latency": 2079.281,
                "stderr": 9.471,
                "cost_per_test": 7.304788,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 62.963,
                "latency": 3586.008,
                "stderr": 9.471,
                "cost_per_test": 25.254544,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 59.259,
                "latency": 3900.655,
                "stderr": 9.636,
                "cost_per_test": 0.854674,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 55.556,
                "latency": 2589.489,
                "stderr": 9.745,
                "cost_per_test": 2.051403,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 51.852,
                "latency": 744.696,
                "stderr": 9.799,
                "cost_per_test": 0.267966,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 51.852,
                "latency": 806.663,
                "stderr": 9.799,
                "cost_per_test": 2.997652,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 51.852,
                "latency": 1965.433,
                "stderr": 9.799,
                "cost_per_test": 6.770674,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 48.148,
                "latency": 308.847,
                "stderr": 9.799,
                "cost_per_test": 0.367043,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 48.148,
                "latency": 663.732,
                "stderr": 9.799,
                "cost_per_test": 0.239442,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 48.148,
                "latency": 2136.958,
                "stderr": 9.799,
                "cost_per_test": 0.917209,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 48.148,
                "latency": 3262.96,
                "stderr": 9.799,
                "cost_per_test": 9.405202,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 48.148,
                "latency": 3478.586,
                "stderr": 9.799,
                "cost_per_test": 2.299147,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 48.148,
                "latency": 5187.901,
                "stderr": 9.799,
                "cost_per_test": 2.080453,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 48.148,
                "latency": 6231.017,
                "stderr": 9.799,
                "cost_per_test": 15.071116,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 48.148,
                "latency": 6638.643,
                "stderr": 9.799,
                "cost_per_test": 22.569939,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 48.148,
                "latency": 7758.778,
                "stderr": 9.799,
                "cost_per_test": 5.531551,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 44.444,
                "latency": 838.559,
                "stderr": 9.745,
                "cost_per_test": 4.229889,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 44.444,
                "latency": 1104.503,
                "stderr": 9.745,
                "cost_per_test": 0.246003,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 44.444,
                "latency": 1439.49,
                "stderr": 9.745,
                "cost_per_test": 5.606604,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 44.444,
                "latency": 1439.733,
                "stderr": 9.745,
                "cost_per_test": 3.415131,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 44.444,
                "latency": 1489.886,
                "stderr": 9.745,
                "cost_per_test": 10.123039,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 44.444,
                "latency": 1956.425,
                "stderr": 9.745,
                "cost_per_test": 3.400541,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 44.444,
                "latency": 2408.394,
                "stderr": 9.745,
                "cost_per_test": 7.00366,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 44.444,
                "latency": 2608.688,
                "stderr": 9.745,
                "cost_per_test": 0.055323,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 40.741,
                "latency": 557.173,
                "stderr": 9.636,
                "cost_per_test": 0.757486,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 40.741,
                "latency": 1743.79,
                "stderr": 9.636,
                "cost_per_test": 13.735474,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 40.741,
                "latency": 2240.6,
                "stderr": 9.636,
                "cost_per_test": 0.207888,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 40.741,
                "latency": 2280.787,
                "stderr": 9.636,
                "cost_per_test": 1.099633,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 40.741,
                "latency": 2417.44,
                "stderr": 9.636,
                "cost_per_test": 0.338953,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 40.741,
                "latency": 2689.198,
                "stderr": 9.636,
                "cost_per_test": 4.546668,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 40.741,
                "latency": 3261.102,
                "stderr": 9.636,
                "cost_per_test": 0.259741,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 40.741,
                "latency": 3816.903,
                "stderr": 9.636,
                "cost_per_test": 18.560919,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 40.741,
                "latency": 5529.804,
                "stderr": 9.636,
                "cost_per_test": 1.750979,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 37.037,
                "latency": 304.101,
                "stderr": 9.471,
                "cost_per_test": 0.602276,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 37.037,
                "latency": 538.692,
                "stderr": 9.471,
                "cost_per_test": 2.902212,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 37.037,
                "latency": 885.474,
                "stderr": 9.471,
                "cost_per_test": 4.274558,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 37.037,
                "latency": 1539.561,
                "stderr": 9.471,
                "cost_per_test": 1.681204,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 37.037,
                "latency": 1554.157,
                "stderr": 9.471,
                "cost_per_test": 1.138521,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 37.037,
                "latency": 2088.431,
                "stderr": 9.471,
                "cost_per_test": 0.658074,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 33.333,
                "latency": 740.554,
                "stderr": 9.245,
                "cost_per_test": 0.559142,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 33.333,
                "latency": 1095.538,
                "stderr": 9.245,
                "cost_per_test": 0.736017,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 33.333,
                "latency": 1225.988,
                "stderr": 9.245,
                "cost_per_test": 1.215179,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 33.333,
                "latency": 5842.182,
                "stderr": 9.245,
                "cost_per_test": 2.843836,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 29.63,
                "latency": 254.44,
                "stderr": 8.955,
                "cost_per_test": 1.35934,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 29.63,
                "latency": 288.591,
                "stderr": 8.955,
                "cost_per_test": 0.719234,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 29.63,
                "latency": 493.563,
                "stderr": 8.955,
                "cost_per_test": 0.076555,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 29.63,
                "latency": 840.078,
                "stderr": 8.955,
                "cost_per_test": 0.059764,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 29.63,
                "latency": 1001.064,
                "stderr": 8.955,
                "cost_per_test": 2.550366,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 29.63,
                "latency": 2447.426,
                "stderr": 8.955,
                "cost_per_test": 3.801126,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 29.63,
                "latency": 3932.081,
                "stderr": 8.955,
                "cost_per_test": 7.563297,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 29.63,
                "latency": 6255.777,
                "stderr": 8.955,
                "cost_per_test": 7.220094,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 25.926,
                "latency": 246.451,
                "stderr": 8.594,
                "cost_per_test": 0.200307,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 25.926,
                "latency": 775.441,
                "stderr": 8.594,
                "cost_per_test": 0.228401,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 25.926,
                "latency": 823.116,
                "stderr": 8.594,
                "cost_per_test": 0.385069,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 25.926,
                "latency": 876.732,
                "stderr": 8.594,
                "cost_per_test": 1.504309,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 25.926,
                "latency": 885.063,
                "stderr": 8.594,
                "cost_per_test": 0.857222,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 25.926,
                "latency": 1273.914,
                "stderr": 8.594,
                "cost_per_test": 1.590079,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 25.926,
                "latency": 1304.611,
                "stderr": 8.594,
                "cost_per_test": 1.737788,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 25.926,
                "latency": 2149.371,
                "stderr": 8.594,
                "cost_per_test": 0.78097,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 25.926,
                "latency": 2593.828,
                "stderr": 8.594,
                "cost_per_test": 0.46119,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 22.222,
                "latency": 92.542,
                "stderr": 8.153,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 22.222,
                "latency": 255.111,
                "stderr": 8.153,
                "cost_per_test": 1.230964,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 22.222,
                "latency": 361.704,
                "stderr": 8.153,
                "cost_per_test": 0.473442,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 22.222,
                "latency": 478.114,
                "stderr": 8.153,
                "cost_per_test": 0.031614,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 22.222,
                "latency": 676.765,
                "stderr": 8.153,
                "cost_per_test": 0.13253,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 22.222,
                "latency": 690.614,
                "stderr": 8.153,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 18.519,
                "latency": 89.721,
                "stderr": 7.618,
                "cost_per_test": 0.06691,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 18.519,
                "latency": 192.418,
                "stderr": 7.618,
                "cost_per_test": 0.477549,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 18.519,
                "latency": 1372.833,
                "stderr": 7.618,
                "cost_per_test": 0.220822,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 18.519,
                "latency": 1441.302,
                "stderr": 7.618,
                "cost_per_test": 5.211283,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 18.519,
                "latency": 1733.848,
                "stderr": 7.618,
                "cost_per_test": 1.440762,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 14.815,
                "latency": 428.141,
                "stderr": 6.967,
                "cost_per_test": 0.242453,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 14.815,
                "latency": 698.535,
                "stderr": 6.967,
                "cost_per_test": 0.632609,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 14.815,
                "latency": 2120.497,
                "stderr": 6.967,
                "cost_per_test": 1.875076,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 14.815,
                "latency": 4095.85,
                "stderr": 6.967,
                "cost_per_test": 3.018691,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 11.111,
                "latency": 56.454,
                "stderr": 6.163,
                "cost_per_test": 0.108873,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 11.111,
                "latency": 215.912,
                "stderr": 6.163,
                "cost_per_test": 0.132078,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 11.111,
                "latency": 225.619,
                "stderr": 6.163,
                "cost_per_test": 0.508937,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 11.111,
                "latency": 416.27,
                "stderr": 6.163,
                "cost_per_test": 0.037073,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 11.111,
                "latency": 486.481,
                "stderr": 6.163,
                "cost_per_test": 0.080617,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 11.111,
                "latency": 595.409,
                "stderr": 6.163,
                "cost_per_test": 1.316462,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 7.407,
                "latency": 277.344,
                "stderr": 5.136,
                "cost_per_test": 0.287416,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 7.407,
                "latency": 725.151,
                "stderr": 5.136,
                "cost_per_test": 0.180106,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 3.704,
                "latency": 19.793,
                "stderr": 3.704,
                "cost_per_test": 0.013273,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 3.704,
                "latency": 47.033,
                "stderr": 3.704,
                "cost_per_test": 0.064712,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 3.704,
                "latency": 68.656,
                "stderr": 3.704,
                "cost_per_test": 0.19461,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 3.704,
                "latency": 273.794,
                "stderr": 3.704,
                "cost_per_test": 0.004442,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 3.704,
                "latency": 332.669,
                "stderr": 3.704,
                "cost_per_test": 0.119391,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 186.413,
                "stderr": 0.0,
                "cost_per_test": 0.040156,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 0.0,
                "latency": 450.642,
                "stderr": 0.0,
                "cost_per_test": 0.229555,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            }
        },
        "constitutional_civil_rights": {
            "google/gemini-4-carbon": {
                "accuracy": 55.263,
                "latency": 2436.37,
                "stderr": 8.174,
                "cost_per_test": 6.991706,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 50.0,
                "latency": 261.358,
                "stderr": 8.22,
                "cost_per_test": 0.485181,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 50.0,
                "latency": 977.356,
                "stderr": 8.22,
                "cost_per_test": 6.898245,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 47.368,
                "latency": 1151.764,
                "stderr": 8.209,
                "cost_per_test": 2.721277,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 47.368,
                "latency": 1361.337,
                "stderr": 8.209,
                "cost_per_test": 4.353414,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 47.368,
                "latency": 1681.88,
                "stderr": 8.209,
                "cost_per_test": 0.17967,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 44.737,
                "latency": 427.276,
                "stderr": 8.174,
                "cost_per_test": 2.058545,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 44.737,
                "latency": 1566.47,
                "stderr": 8.174,
                "cost_per_test": 2.477637,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 44.737,
                "latency": 1835.592,
                "stderr": 8.174,
                "cost_per_test": 0.898996,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 44.737,
                "latency": 1898.737,
                "stderr": 8.174,
                "cost_per_test": 6.396226,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 44.737,
                "latency": 2446.246,
                "stderr": 8.174,
                "cost_per_test": 0.179145,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 44.737,
                "latency": 5507.901,
                "stderr": 8.174,
                "cost_per_test": 1.55544,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 44.737,
                "latency": 6953.006,
                "stderr": 8.174,
                "cost_per_test": 4.449226,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 42.105,
                "latency": 220.844,
                "stderr": 8.117,
                "cost_per_test": 1.043129,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 42.105,
                "latency": 1049.066,
                "stderr": 8.117,
                "cost_per_test": 4.00044,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 42.105,
                "latency": 1675.051,
                "stderr": 8.117,
                "cost_per_test": 0.53415,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 42.105,
                "latency": 1748.11,
                "stderr": 8.117,
                "cost_per_test": 2.722205,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 42.105,
                "latency": 3510.897,
                "stderr": 8.117,
                "cost_per_test": 21.64278,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 42.105,
                "latency": 5827.977,
                "stderr": 8.117,
                "cost_per_test": 14.117789,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 42.105,
                "latency": 6438.62,
                "stderr": 8.117,
                "cost_per_test": 22.403219,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 39.474,
                "latency": 275.608,
                "stderr": 8.036,
                "cost_per_test": 0.430216,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 39.474,
                "latency": 609.116,
                "stderr": 8.036,
                "cost_per_test": 2.462184,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 39.474,
                "latency": 632.775,
                "stderr": 8.036,
                "cost_per_test": 2.959097,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 39.474,
                "latency": 645.429,
                "stderr": 8.036,
                "cost_per_test": 0.225678,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 39.474,
                "latency": 1943.878,
                "stderr": 8.036,
                "cost_per_test": 0.828943,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 39.474,
                "latency": 2063.284,
                "stderr": 8.036,
                "cost_per_test": 0.037714,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 39.474,
                "latency": 2583.919,
                "stderr": 8.036,
                "cost_per_test": 1.779991,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 39.474,
                "latency": 2858.564,
                "stderr": 8.036,
                "cost_per_test": 0.671944,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 36.842,
                "latency": 518.106,
                "stderr": 7.93,
                "cost_per_test": 0.191689,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 36.842,
                "latency": 572.241,
                "stderr": 7.93,
                "cost_per_test": 2.490256,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 36.842,
                "latency": 1396.958,
                "stderr": 7.93,
                "cost_per_test": 1.277781,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 36.842,
                "latency": 1461.512,
                "stderr": 7.93,
                "cost_per_test": 0.943808,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 34.211,
                "latency": 353.782,
                "stderr": 7.799,
                "cost_per_test": 0.052994,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 34.211,
                "latency": 355.395,
                "stderr": 7.799,
                "cost_per_test": 0.362114,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 34.211,
                "latency": 522.986,
                "stderr": 7.799,
                "cost_per_test": 0.545149,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 34.211,
                "latency": 733.352,
                "stderr": 7.799,
                "cost_per_test": 0.291222,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 34.211,
                "latency": 830.725,
                "stderr": 7.799,
                "cost_per_test": 0.076515,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 34.211,
                "latency": 853.113,
                "stderr": 7.799,
                "cost_per_test": 1.965994,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 34.211,
                "latency": 1718.661,
                "stderr": 7.799,
                "cost_per_test": 0.583776,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 34.211,
                "latency": 1785.016,
                "stderr": 7.799,
                "cost_per_test": 0.721618,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 34.211,
                "latency": 3263.997,
                "stderr": 7.799,
                "cost_per_test": 2.246783,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 34.211,
                "latency": 3655.679,
                "stderr": 7.799,
                "cost_per_test": 6.783177,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 34.211,
                "latency": 3868.674,
                "stderr": 7.799,
                "cost_per_test": 14.526185,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 34.211,
                "latency": 4486.623,
                "stderr": 7.799,
                "cost_per_test": 1.742342,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 34.211,
                "latency": 4751.209,
                "stderr": 7.799,
                "cost_per_test": 2.624566,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 31.579,
                "latency": 62.09,
                "stderr": 7.642,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 31.579,
                "latency": 214.833,
                "stderr": 7.642,
                "cost_per_test": 0.667444,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 31.579,
                "latency": 611.367,
                "stderr": 7.642,
                "cost_per_test": 0.372278,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 31.579,
                "latency": 907.475,
                "stderr": 7.642,
                "cost_per_test": 0.836603,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 31.579,
                "latency": 1025.06,
                "stderr": 7.642,
                "cost_per_test": 0.204391,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 31.579,
                "latency": 1775.958,
                "stderr": 7.642,
                "cost_per_test": 2.692913,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 31.579,
                "latency": 1958.935,
                "stderr": 7.642,
                "cost_per_test": 0.364867,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 31.579,
                "latency": 6612.654,
                "stderr": 7.642,
                "cost_per_test": 7.25464,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 28.947,
                "latency": 224.844,
                "stderr": 7.456,
                "cost_per_test": 0.987781,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 28.947,
                "latency": 310.439,
                "stderr": 7.456,
                "cost_per_test": 0.303515,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 28.947,
                "latency": 558.667,
                "stderr": 7.456,
                "cost_per_test": 0.395873,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 28.947,
                "latency": 2156.642,
                "stderr": 7.456,
                "cost_per_test": 5.731692,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 28.947,
                "latency": 2834.861,
                "stderr": 7.456,
                "cost_per_test": 0.421978,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 26.316,
                "latency": 1280.117,
                "stderr": 7.239,
                "cost_per_test": 8.37428,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 26.316,
                "latency": 1286.448,
                "stderr": 7.239,
                "cost_per_test": 1.563366,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 23.684,
                "latency": 782.498,
                "stderr": 6.989,
                "cost_per_test": 1.136745,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 23.684,
                "latency": 3732.629,
                "stderr": 6.989,
                "cost_per_test": 2.673236,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 21.053,
                "latency": 61.408,
                "stderr": 6.702,
                "cost_per_test": 0.080105,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 21.053,
                "latency": 223.214,
                "stderr": 6.702,
                "cost_per_test": 0.466165,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 21.053,
                "latency": 828.518,
                "stderr": 6.702,
                "cost_per_test": 0.208128,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 21.053,
                "latency": 1564.212,
                "stderr": 6.702,
                "cost_per_test": 1.25835,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 18.421,
                "latency": 160.664,
                "stderr": 6.373,
                "cost_per_test": 0.393807,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 18.421,
                "latency": 279.901,
                "stderr": 6.373,
                "cost_per_test": 0.255064,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 18.421,
                "latency": 1202.973,
                "stderr": 6.373,
                "cost_per_test": 4.442219,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 15.789,
                "latency": 1326.72,
                "stderr": 5.995,
                "cost_per_test": 0.216972,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 13.158,
                "latency": 74.98,
                "stderr": 5.557,
                "cost_per_test": 0.047883,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 13.158,
                "latency": 125.212,
                "stderr": 5.557,
                "cost_per_test": 0.250093,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 13.158,
                "latency": 276.602,
                "stderr": 5.557,
                "cost_per_test": 0.260695,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 13.158,
                "latency": 384.09,
                "stderr": 5.557,
                "cost_per_test": 0.079592,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 13.158,
                "latency": 822.903,
                "stderr": 5.557,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 13.158,
                "latency": 1096.62,
                "stderr": 5.557,
                "cost_per_test": 0.614384,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 13.158,
                "latency": 1433.795,
                "stderr": 5.557,
                "cost_per_test": 1.467472,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 10.526,
                "latency": 303.387,
                "stderr": 5.045,
                "cost_per_test": 0.019445,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 10.526,
                "latency": 449.168,
                "stderr": 5.045,
                "cost_per_test": 0.209051,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 10.526,
                "latency": 589.79,
                "stderr": 5.045,
                "cost_per_test": 0.111862,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 10.526,
                "latency": 696.299,
                "stderr": 5.045,
                "cost_per_test": 0.516518,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 10.526,
                "latency": 715.501,
                "stderr": 5.045,
                "cost_per_test": 1.425655,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 7.895,
                "latency": 214.461,
                "stderr": 4.433,
                "cost_per_test": 0.119798,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 7.895,
                "latency": 352.653,
                "stderr": 4.433,
                "cost_per_test": 0.029482,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 5.263,
                "latency": 268.813,
                "stderr": 3.671,
                "cost_per_test": 0.003527,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 2.632,
                "latency": 21.603,
                "stderr": 2.632,
                "cost_per_test": 0.013343,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 2.632,
                "latency": 53.037,
                "stderr": 2.632,
                "cost_per_test": 0.058844,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 2.632,
                "latency": 540.649,
                "stderr": 2.632,
                "cost_per_test": 0.230889,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 2.632,
                "latency": 597.509,
                "stderr": 2.632,
                "cost_per_test": 0.134636,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 176.967,
                "stderr": 0.0,
                "cost_per_test": 0.030415,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.0,
                "latency": 293.416,
                "stderr": 0.0,
                "cost_per_test": 0.090182,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            }
        },
        "criminal": {
            "google/gemini-4-carbon": {
                "accuracy": 59.649,
                "latency": 3164.955,
                "stderr": 6.556,
                "cost_per_test": 8.842662,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 56.14,
                "latency": 1907.88,
                "stderr": 6.631,
                "cost_per_test": 6.616583,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 49.123,
                "latency": 1527.247,
                "stderr": 6.681,
                "cost_per_test": 3.617802,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 47.368,
                "latency": 303.011,
                "stderr": 6.672,
                "cost_per_test": 0.60538,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 47.368,
                "latency": 1371.988,
                "stderr": 6.672,
                "cost_per_test": 5.268708,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 45.614,
                "latency": 755.757,
                "stderr": 6.656,
                "cost_per_test": 2.852274,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 43.86,
                "latency": 967.654,
                "stderr": 6.631,
                "cost_per_test": 4.059119,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 43.86,
                "latency": 1928.013,
                "stderr": 6.631,
                "cost_per_test": 0.177786,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 43.86,
                "latency": 2118.512,
                "stderr": 6.631,
                "cost_per_test": 6.903486,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 43.86,
                "latency": 4987.986,
                "stderr": 6.631,
                "cost_per_test": 24.12378,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 43.86,
                "latency": 7534.426,
                "stderr": 6.631,
                "cost_per_test": 5.274524,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 42.105,
                "latency": 1356.005,
                "stderr": 6.598,
                "cost_per_test": 1.460147,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 42.105,
                "latency": 1398.537,
                "stderr": 6.598,
                "cost_per_test": 9.468125,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 42.105,
                "latency": 5566.578,
                "stderr": 6.598,
                "cost_per_test": 1.725041,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 42.105,
                "latency": 7162.351,
                "stderr": 6.598,
                "cost_per_test": 25.272629,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 40.351,
                "latency": 315.871,
                "stderr": 6.556,
                "cost_per_test": 1.534611,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 40.351,
                "latency": 1506.644,
                "stderr": 6.556,
                "cost_per_test": 2.518676,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 40.351,
                "latency": 3997.963,
                "stderr": 6.556,
                "cost_per_test": 2.676114,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 38.596,
                "latency": 301.211,
                "stderr": 6.505,
                "cost_per_test": 0.446419,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 38.596,
                "latency": 2210.107,
                "stderr": 6.505,
                "cost_per_test": 0.87429,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 38.596,
                "latency": 2957.873,
                "stderr": 6.505,
                "cost_per_test": 2.182135,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 38.596,
                "latency": 3466.552,
                "stderr": 6.505,
                "cost_per_test": 22.326482,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 38.596,
                "latency": 6259.782,
                "stderr": 6.505,
                "cost_per_test": 14.876209,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 36.842,
                "latency": 320.962,
                "stderr": 6.446,
                "cost_per_test": 0.345173,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 36.842,
                "latency": 767.942,
                "stderr": 6.446,
                "cost_per_test": 0.321616,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 36.842,
                "latency": 772.67,
                "stderr": 6.446,
                "cost_per_test": 3.543085,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 36.842,
                "latency": 874.891,
                "stderr": 6.446,
                "cost_per_test": 0.822469,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 36.842,
                "latency": 989.26,
                "stderr": 6.446,
                "cost_per_test": 1.99149,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 36.842,
                "latency": 2310.474,
                "stderr": 6.446,
                "cost_per_test": 7.322657,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 36.842,
                "latency": 3744.173,
                "stderr": 6.446,
                "cost_per_test": 0.819373,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 35.088,
                "latency": 540.381,
                "stderr": 6.377,
                "cost_per_test": 0.616709,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 35.088,
                "latency": 2183.678,
                "stderr": 6.377,
                "cost_per_test": 2.568801,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 35.088,
                "latency": 2208.959,
                "stderr": 6.377,
                "cost_per_test": 0.046529,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 35.088,
                "latency": 2224.103,
                "stderr": 6.377,
                "cost_per_test": 0.715843,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 35.088,
                "latency": 2341.734,
                "stderr": 6.377,
                "cost_per_test": 3.38256,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 35.088,
                "latency": 2498.94,
                "stderr": 6.377,
                "cost_per_test": 1.017486,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 35.088,
                "latency": 4543.691,
                "stderr": 6.377,
                "cost_per_test": 8.912211,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 33.333,
                "latency": 347.571,
                "stderr": 6.299,
                "cost_per_test": 0.827264,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 33.333,
                "latency": 682.99,
                "stderr": 6.299,
                "cost_per_test": 0.233636,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 33.333,
                "latency": 1712.774,
                "stderr": 6.299,
                "cost_per_test": 1.169956,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 33.333,
                "latency": 2749.369,
                "stderr": 6.299,
                "cost_per_test": 0.250408,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 33.333,
                "latency": 5104.912,
                "stderr": 6.299,
                "cost_per_test": 2.104097,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 31.579,
                "latency": 2079.561,
                "stderr": 6.212,
                "cost_per_test": 3.601082,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 31.579,
                "latency": 5432.346,
                "stderr": 6.212,
                "cost_per_test": 3.124013,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 29.825,
                "latency": 331.685,
                "stderr": 6.113,
                "cost_per_test": 0.405872,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 29.825,
                "latency": 425.103,
                "stderr": 6.113,
                "cost_per_test": 0.06767,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 29.825,
                "latency": 529.681,
                "stderr": 6.113,
                "cost_per_test": 2.779258,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 29.825,
                "latency": 1025.226,
                "stderr": 6.113,
                "cost_per_test": 0.071477,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 29.825,
                "latency": 1658.26,
                "stderr": 6.113,
                "cost_per_test": 10.941763,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 29.825,
                "latency": 2563.313,
                "stderr": 6.113,
                "cost_per_test": 0.994665,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 28.07,
                "latency": 6680.114,
                "stderr": 6.005,
                "cost_per_test": 8.328752,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 26.316,
                "latency": 256.075,
                "stderr": 5.884,
                "cost_per_test": 1.359179,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 26.316,
                "latency": 855.631,
                "stderr": 5.884,
                "cost_per_test": 0.325282,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 26.316,
                "latency": 1433.199,
                "stderr": 5.884,
                "cost_per_test": 1.995877,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 26.316,
                "latency": 1562.595,
                "stderr": 5.884,
                "cost_per_test": 5.495004,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 26.316,
                "latency": 2463.873,
                "stderr": 5.884,
                "cost_per_test": 0.454694,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 26.316,
                "latency": 2805.101,
                "stderr": 5.884,
                "cost_per_test": 0.425544,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 24.561,
                "latency": 189.608,
                "stderr": 5.752,
                "cost_per_test": 0.501214,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 24.561,
                "latency": 711.836,
                "stderr": 5.752,
                "cost_per_test": 0.143249,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 24.561,
                "latency": 1264.855,
                "stderr": 5.752,
                "cost_per_test": 0.25224,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 22.807,
                "latency": 1034.653,
                "stderr": 5.607,
                "cost_per_test": 0.846458,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 21.053,
                "latency": 787.984,
                "stderr": 5.448,
                "cost_per_test": 1.332257,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 21.053,
                "latency": 872.79,
                "stderr": 5.448,
                "cost_per_test": 0.509253,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 19.298,
                "latency": 87.152,
                "stderr": 5.274,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 19.298,
                "latency": 217.466,
                "stderr": 5.274,
                "cost_per_test": 0.48486,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 17.544,
                "latency": 267.815,
                "stderr": 5.083,
                "cost_per_test": 0.155562,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 17.544,
                "latency": 4308.911,
                "stderr": 5.083,
                "cost_per_test": 3.298261,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 15.789,
                "latency": 651.134,
                "stderr": 4.873,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 15.789,
                "latency": 858.129,
                "stderr": 4.873,
                "cost_per_test": 0.2621,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 15.789,
                "latency": 1004.085,
                "stderr": 4.873,
                "cost_per_test": 0.615927,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 15.789,
                "latency": 1398.791,
                "stderr": 4.873,
                "cost_per_test": 0.25486,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 14.035,
                "latency": 110.22,
                "stderr": 4.642,
                "cost_per_test": 0.199087,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 14.035,
                "latency": 398.106,
                "stderr": 4.642,
                "cost_per_test": 0.081998,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 14.035,
                "latency": 1937.218,
                "stderr": 4.642,
                "cost_per_test": 1.608029,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 12.281,
                "latency": 63.499,
                "stderr": 4.386,
                "cost_per_test": 0.122864,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 12.281,
                "latency": 221.166,
                "stderr": 4.386,
                "cost_per_test": 0.13559,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 12.281,
                "latency": 435.186,
                "stderr": 4.386,
                "cost_per_test": 0.022099,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 10.526,
                "latency": 283.18,
                "stderr": 4.101,
                "cost_per_test": 0.260687,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 10.526,
                "latency": 611.807,
                "stderr": 4.101,
                "cost_per_test": 1.490522,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 10.526,
                "latency": 698.746,
                "stderr": 4.101,
                "cost_per_test": 0.68798,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 8.772,
                "latency": 78.654,
                "stderr": 3.78,
                "cost_per_test": 0.062754,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 7.018,
                "latency": 394.953,
                "stderr": 3.413,
                "cost_per_test": 0.023143,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 5.263,
                "latency": 421.649,
                "stderr": 2.984,
                "cost_per_test": 0.226204,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 5.263,
                "latency": 2330.292,
                "stderr": 2.984,
                "cost_per_test": 2.024329,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 3.509,
                "latency": 187.227,
                "stderr": 2.459,
                "cost_per_test": 0.003839,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 3.509,
                "latency": 548.936,
                "stderr": 2.459,
                "cost_per_test": 0.153371,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 1.754,
                "latency": 20.102,
                "stderr": 1.754,
                "cost_per_test": 0.013186,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 1.754,
                "latency": 48.522,
                "stderr": 1.754,
                "cost_per_test": 0.059734,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 185.804,
                "stderr": 0.0,
                "cost_per_test": 0.034171,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.0,
                "latency": 301.868,
                "stderr": 0.0,
                "cost_per_test": 0.108798,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 0.0,
                "latency": 521.807,
                "stderr": 0.0,
                "cost_per_test": 0.26163,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            }
        },
        "family": {
            "google/gemini-4-carbon": {
                "accuracy": 45.455,
                "latency": 4551.243,
                "stderr": 10.866,
                "cost_per_test": 12.648487,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 40.909,
                "latency": 357.472,
                "stderr": 10.729,
                "cost_per_test": 0.633898,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 40.909,
                "latency": 1858.312,
                "stderr": 10.729,
                "cost_per_test": 6.195885,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 40.909,
                "latency": 2827.87,
                "stderr": 10.729,
                "cost_per_test": 0.051257,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 36.364,
                "latency": 1437.652,
                "stderr": 10.497,
                "cost_per_test": 5.323844,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 36.364,
                "latency": 1643.011,
                "stderr": 10.497,
                "cost_per_test": 11.011763,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 36.364,
                "latency": 2821.52,
                "stderr": 10.497,
                "cost_per_test": 10.91868,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 36.364,
                "latency": 8086.246,
                "stderr": 10.497,
                "cost_per_test": 5.806978,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 31.818,
                "latency": 638.032,
                "stderr": 10.164,
                "cost_per_test": 3.584147,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 31.818,
                "latency": 1329.086,
                "stderr": 10.164,
                "cost_per_test": 0.070993,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 31.818,
                "latency": 2250.577,
                "stderr": 10.164,
                "cost_per_test": 4.730031,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 31.818,
                "latency": 3235.665,
                "stderr": 10.164,
                "cost_per_test": 21.529237,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 31.818,
                "latency": 3849.877,
                "stderr": 10.164,
                "cost_per_test": 2.697912,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 31.818,
                "latency": 4670.091,
                "stderr": 10.164,
                "cost_per_test": 0.970682,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 27.273,
                "latency": 328.141,
                "stderr": 9.719,
                "cost_per_test": 0.452803,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 27.273,
                "latency": 716.66,
                "stderr": 9.719,
                "cost_per_test": 0.822081,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 27.273,
                "latency": 1189.264,
                "stderr": 9.719,
                "cost_per_test": 2.408033,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 27.273,
                "latency": 2445.946,
                "stderr": 9.719,
                "cost_per_test": 0.826784,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 27.273,
                "latency": 2649.862,
                "stderr": 9.719,
                "cost_per_test": 0.231872,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 27.273,
                "latency": 3035.777,
                "stderr": 9.719,
                "cost_per_test": 5.439212,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 27.273,
                "latency": 3202.004,
                "stderr": 9.719,
                "cost_per_test": 0.286053,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 27.273,
                "latency": 6801.677,
                "stderr": 9.719,
                "cost_per_test": 2.756433,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 22.727,
                "latency": 348.059,
                "stderr": 9.145,
                "cost_per_test": 0.406411,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 22.727,
                "latency": 855.894,
                "stderr": 9.145,
                "cost_per_test": 0.383913,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 22.727,
                "latency": 966.329,
                "stderr": 9.145,
                "cost_per_test": 4.336354,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 22.727,
                "latency": 1405.167,
                "stderr": 9.145,
                "cost_per_test": 1.425494,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 22.727,
                "latency": 1713.026,
                "stderr": 9.145,
                "cost_per_test": 2.549692,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 22.727,
                "latency": 1806.747,
                "stderr": 9.145,
                "cost_per_test": 1.50639,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 22.727,
                "latency": 2019.893,
                "stderr": 9.145,
                "cost_per_test": 0.418234,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 22.727,
                "latency": 3456.123,
                "stderr": 9.145,
                "cost_per_test": 1.536833,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 22.727,
                "latency": 5152.106,
                "stderr": 9.145,
                "cost_per_test": 1.724456,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 22.727,
                "latency": 7441.78,
                "stderr": 9.145,
                "cost_per_test": 25.835741,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 18.182,
                "latency": 121.037,
                "stderr": 8.417,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 18.182,
                "latency": 306.75,
                "stderr": 8.417,
                "cost_per_test": 1.594003,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 18.182,
                "latency": 613.145,
                "stderr": 8.417,
                "cost_per_test": 2.522005,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 18.182,
                "latency": 787.574,
                "stderr": 8.417,
                "cost_per_test": 2.857814,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 18.182,
                "latency": 1482.157,
                "stderr": 8.417,
                "cost_per_test": 5.56694,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 18.182,
                "latency": 1564.903,
                "stderr": 8.417,
                "cost_per_test": 2.758903,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 18.182,
                "latency": 1762.575,
                "stderr": 8.417,
                "cost_per_test": 0.33418,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 18.182,
                "latency": 1896.183,
                "stderr": 8.417,
                "cost_per_test": 15.003346,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 18.182,
                "latency": 2357.557,
                "stderr": 8.417,
                "cost_per_test": 3.772956,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 18.182,
                "latency": 3017.714,
                "stderr": 8.417,
                "cost_per_test": 1.168665,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 18.182,
                "latency": 4757.56,
                "stderr": 8.417,
                "cost_per_test": 3.063697,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 18.182,
                "latency": 5063.727,
                "stderr": 8.417,
                "cost_per_test": 9.812737,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 18.182,
                "latency": 5320.428,
                "stderr": 8.417,
                "cost_per_test": 12.577804,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 13.636,
                "latency": 390.844,
                "stderr": 7.489,
                "cost_per_test": 0.986437,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 13.636,
                "latency": 951.192,
                "stderr": 7.489,
                "cost_per_test": 0.370092,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 13.636,
                "latency": 1013.398,
                "stderr": 7.489,
                "cost_per_test": 0.323772,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 13.636,
                "latency": 1101.097,
                "stderr": 7.489,
                "cost_per_test": 0.562977,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 13.636,
                "latency": 2332.251,
                "stderr": 7.489,
                "cost_per_test": 1.856297,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 13.636,
                "latency": 2680.585,
                "stderr": 7.489,
                "cost_per_test": 7.622096,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 13.636,
                "latency": 3983.979,
                "stderr": 7.489,
                "cost_per_test": 0.672496,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 13.636,
                "latency": 6569.663,
                "stderr": 7.489,
                "cost_per_test": 30.084962,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 13.636,
                "latency": 6905.045,
                "stderr": 7.489,
                "cost_per_test": 8.216145,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 9.091,
                "latency": 195.123,
                "stderr": 6.273,
                "cost_per_test": 0.581541,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 9.091,
                "latency": 342.126,
                "stderr": 6.273,
                "cost_per_test": 0.41232,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 9.091,
                "latency": 514.571,
                "stderr": 6.273,
                "cost_per_test": 0.08758,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 9.091,
                "latency": 839.919,
                "stderr": 6.273,
                "cost_per_test": 0.991333,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 9.091,
                "latency": 900.57,
                "stderr": 6.273,
                "cost_per_test": 1.420719,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 9.091,
                "latency": 1317.912,
                "stderr": 6.273,
                "cost_per_test": 0.816394,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 9.091,
                "latency": 1722.37,
                "stderr": 6.273,
                "cost_per_test": 0.30233,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 9.091,
                "latency": 1758.115,
                "stderr": 6.273,
                "cost_per_test": 5.389618,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 9.091,
                "latency": 2391.966,
                "stderr": 6.273,
                "cost_per_test": 0.88974,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 9.091,
                "latency": 2917.375,
                "stderr": 6.273,
                "cost_per_test": 0.465031,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 9.091,
                "latency": 6312.514,
                "stderr": 6.273,
                "cost_per_test": 3.697863,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 4.545,
                "latency": 62.256,
                "stderr": 4.545,
                "cost_per_test": 0.169797,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 4.545,
                "latency": 98.358,
                "stderr": 4.545,
                "cost_per_test": 0.074199,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 4.545,
                "latency": 167.284,
                "stderr": 4.545,
                "cost_per_test": 0.168183,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 4.545,
                "latency": 434.377,
                "stderr": 4.545,
                "cost_per_test": 4.138935,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 4.545,
                "latency": 435.98,
                "stderr": 4.545,
                "cost_per_test": 0.075731,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 4.545,
                "latency": 458.965,
                "stderr": 4.545,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 4.545,
                "latency": 692.635,
                "stderr": 4.545,
                "cost_per_test": 0.039086,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 4.545,
                "latency": 1001.225,
                "stderr": 4.545,
                "cost_per_test": 0.173654,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 4.545,
                "latency": 4690.594,
                "stderr": 4.545,
                "cost_per_test": 3.506565,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 0.0,
                "latency": 18.535,
                "stderr": 0.0,
                "cost_per_test": 0.013412,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 0.0,
                "latency": 47.19,
                "stderr": 0.0,
                "cost_per_test": 0.062993,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 0.0,
                "latency": 86.589,
                "stderr": 0.0,
                "cost_per_test": 0.230167,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 222.566,
                "stderr": 0.0,
                "cost_per_test": 0.036939,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 0.0,
                "latency": 246.131,
                "stderr": 0.0,
                "cost_per_test": 0.16395,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 0.0,
                "latency": 250.897,
                "stderr": 0.0,
                "cost_per_test": 0.288946,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 0.0,
                "latency": 311.615,
                "stderr": 0.0,
                "cost_per_test": 0.005312,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.0,
                "latency": 383.551,
                "stderr": 0.0,
                "cost_per_test": 0.13236,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 0.0,
                "latency": 422.665,
                "stderr": 0.0,
                "cost_per_test": 0.030982,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 0.0,
                "latency": 436.535,
                "stderr": 0.0,
                "cost_per_test": 0.255715,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 0.0,
                "latency": 532.251,
                "stderr": 0.0,
                "cost_per_test": 0.261764,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 0.0,
                "latency": 570.598,
                "stderr": 0.0,
                "cost_per_test": 1.090579,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 0.0,
                "latency": 592.178,
                "stderr": 0.0,
                "cost_per_test": 0.140683,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 0.0,
                "latency": 727.744,
                "stderr": 0.0,
                "cost_per_test": 1.001303,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 0.0,
                "latency": 1055.917,
                "stderr": 0.0,
                "cost_per_test": 0.443026,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 0.0,
                "latency": 2863.795,
                "stderr": 0.0,
                "cost_per_test": 2.335677,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 0.0,
                "latency": 4205.69,
                "stderr": 0.0,
                "cost_per_test": 2.933958,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            }
        },
        "health": {
            "fireworks/ember-1": {
                "accuracy": 80.0,
                "latency": 2538.105,
                "stderr": 7.428,
                "cost_per_test": 4.232271,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 80.0,
                "latency": 3030.813,
                "stderr": 7.428,
                "cost_per_test": 22.907678,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 80.0,
                "latency": 3850.231,
                "stderr": 7.428,
                "cost_per_test": 8.132992,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 80.0,
                "latency": 4151.253,
                "stderr": 7.428,
                "cost_per_test": 22.836198,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 80.0,
                "latency": 4630.853,
                "stderr": 7.428,
                "cost_per_test": 12.659413,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 80.0,
                "latency": 5675.961,
                "stderr": 7.428,
                "cost_per_test": 20.596615,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 76.667,
                "latency": 836.886,
                "stderr": 7.854,
                "cost_per_test": 4.339603,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 76.667,
                "latency": 858.216,
                "stderr": 7.854,
                "cost_per_test": 4.579894,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 76.667,
                "latency": 986.856,
                "stderr": 7.854,
                "cost_per_test": 10.254175,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 76.667,
                "latency": 1095.281,
                "stderr": 7.854,
                "cost_per_test": 4.723349,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 76.667,
                "latency": 1338.036,
                "stderr": 7.854,
                "cost_per_test": 4.625213,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 76.667,
                "latency": 1346.948,
                "stderr": 7.854,
                "cost_per_test": 7.009258,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 76.667,
                "latency": 1414.958,
                "stderr": 7.854,
                "cost_per_test": 3.858235,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 76.667,
                "latency": 3215.749,
                "stderr": 7.854,
                "cost_per_test": 2.470231,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 73.333,
                "latency": 318.119,
                "stderr": 8.212,
                "cost_per_test": 0.64171,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 73.333,
                "latency": 706.355,
                "stderr": 8.212,
                "cost_per_test": 0.272236,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 73.333,
                "latency": 1693.184,
                "stderr": 8.212,
                "cost_per_test": 1.828426,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 73.333,
                "latency": 3278.164,
                "stderr": 8.212,
                "cost_per_test": 0.238934,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 73.333,
                "latency": 3496.535,
                "stderr": 8.212,
                "cost_per_test": 2.5406,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 73.333,
                "latency": 5711.873,
                "stderr": 8.212,
                "cost_per_test": 6.862578,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 70.0,
                "latency": 470.162,
                "stderr": 8.51,
                "cost_per_test": 0.708927,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 70.0,
                "latency": 710.57,
                "stderr": 8.51,
                "cost_per_test": 0.327988,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 70.0,
                "latency": 1360.302,
                "stderr": 8.51,
                "cost_per_test": 0.193071,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 70.0,
                "latency": 2414.0,
                "stderr": 8.51,
                "cost_per_test": 0.809076,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 70.0,
                "latency": 2450.538,
                "stderr": 8.51,
                "cost_per_test": 6.751201,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 66.667,
                "latency": 1126.576,
                "stderr": 8.754,
                "cost_per_test": 12.38182,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 66.667,
                "latency": 1266.05,
                "stderr": 8.754,
                "cost_per_test": 2.204201,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 66.667,
                "latency": 1553.826,
                "stderr": 8.754,
                "cost_per_test": 0.926438,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 66.667,
                "latency": 1838.285,
                "stderr": 8.754,
                "cost_per_test": 3.207198,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 66.667,
                "latency": 1923.984,
                "stderr": 8.754,
                "cost_per_test": 0.708547,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 66.667,
                "latency": 2062.802,
                "stderr": 8.754,
                "cost_per_test": 9.360291,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 66.667,
                "latency": 3508.124,
                "stderr": 8.754,
                "cost_per_test": 1.530774,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 66.667,
                "latency": 3604.391,
                "stderr": 8.754,
                "cost_per_test": 0.88574,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 63.333,
                "latency": 1364.3,
                "stderr": 8.949,
                "cost_per_test": 2.657382,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 63.333,
                "latency": 1922.826,
                "stderr": 8.949,
                "cost_per_test": 2.642852,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 63.333,
                "latency": 5571.399,
                "stderr": 8.949,
                "cost_per_test": 2.397715,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 60.0,
                "latency": 1026.417,
                "stderr": 9.097,
                "cost_per_test": 3.793295,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 60.0,
                "latency": 2681.332,
                "stderr": 9.097,
                "cost_per_test": 0.057691,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 60.0,
                "latency": 2705.795,
                "stderr": 9.097,
                "cost_per_test": 1.314815,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 56.667,
                "latency": 282.896,
                "stderr": 9.202,
                "cost_per_test": 1.899416,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 56.667,
                "latency": 303.781,
                "stderr": 9.202,
                "cost_per_test": 0.574895,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 56.667,
                "latency": 3359.798,
                "stderr": 9.202,
                "cost_per_test": 3.11111,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 56.667,
                "latency": 7418.628,
                "stderr": 9.202,
                "cost_per_test": 5.193942,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 53.333,
                "latency": 462.415,
                "stderr": 9.264,
                "cost_per_test": 4.040025,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 53.333,
                "latency": 663.425,
                "stderr": 9.264,
                "cost_per_test": 3.159294,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 53.333,
                "latency": 781.127,
                "stderr": 9.264,
                "cost_per_test": 0.081431,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 53.333,
                "latency": 1047.406,
                "stderr": 9.264,
                "cost_per_test": 3.45972,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 53.333,
                "latency": 1685.594,
                "stderr": 9.264,
                "cost_per_test": 1.561613,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 53.333,
                "latency": 2028.962,
                "stderr": 9.264,
                "cost_per_test": 0.905473,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 53.333,
                "latency": 2156.169,
                "stderr": 9.264,
                "cost_per_test": 0.374147,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 50.0,
                "latency": 245.802,
                "stderr": 9.285,
                "cost_per_test": 0.701623,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 50.0,
                "latency": 931.988,
                "stderr": 9.285,
                "cost_per_test": 0.931196,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 50.0,
                "latency": 5897.488,
                "stderr": 9.285,
                "cost_per_test": 2.500509,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 46.667,
                "latency": 97.234,
                "stderr": 9.264,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 46.667,
                "latency": 472.047,
                "stderr": 9.264,
                "cost_per_test": 0.076792,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 46.667,
                "latency": 800.16,
                "stderr": 9.264,
                "cost_per_test": 0.394232,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 46.667,
                "latency": 999.255,
                "stderr": 9.264,
                "cost_per_test": 0.81017,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 43.333,
                "latency": 274.578,
                "stderr": 9.202,
                "cost_per_test": 1.402227,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 43.333,
                "latency": 910.936,
                "stderr": 9.202,
                "cost_per_test": 0.496174,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 40.0,
                "latency": 79.862,
                "stderr": 9.097,
                "cost_per_test": 0.233416,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 40.0,
                "latency": 298.421,
                "stderr": 9.097,
                "cost_per_test": 0.506452,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 40.0,
                "latency": 848.765,
                "stderr": 9.097,
                "cost_per_test": 1.786099,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 36.667,
                "latency": 134.685,
                "stderr": 8.949,
                "cost_per_test": 0.416767,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 36.667,
                "latency": 322.342,
                "stderr": 8.949,
                "cost_per_test": 0.321248,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 36.667,
                "latency": 1384.623,
                "stderr": 8.949,
                "cost_per_test": 0.400672,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 36.667,
                "latency": 3297.206,
                "stderr": 8.949,
                "cost_per_test": 0.553927,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 33.333,
                "latency": 232.285,
                "stderr": 8.754,
                "cost_per_test": 1.804209,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 30.0,
                "latency": 145.057,
                "stderr": 8.51,
                "cost_per_test": 0.136684,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 30.0,
                "latency": 696.803,
                "stderr": 8.51,
                "cost_per_test": 0.848896,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 30.0,
                "latency": 998.171,
                "stderr": 8.51,
                "cost_per_test": 0.155212,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 30.0,
                "latency": 1608.673,
                "stderr": 8.51,
                "cost_per_test": 1.371176,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 26.667,
                "latency": 477.359,
                "stderr": 8.212,
                "cost_per_test": 0.028475,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 26.667,
                "latency": 830.651,
                "stderr": 8.212,
                "cost_per_test": 0.505144,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 23.333,
                "latency": 387.77,
                "stderr": 7.854,
                "cost_per_test": 0.119406,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 16.667,
                "latency": 538.653,
                "stderr": 6.92,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 16.667,
                "latency": 1671.994,
                "stderr": 6.92,
                "cost_per_test": 1.329631,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 13.333,
                "latency": 48.195,
                "stderr": 6.312,
                "cost_per_test": 0.13408,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 13.333,
                "latency": 334.244,
                "stderr": 6.312,
                "cost_per_test": 0.106495,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 13.333,
                "latency": 587.499,
                "stderr": 6.312,
                "cost_per_test": 0.21973,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 10.0,
                "latency": 71.497,
                "stderr": 5.571,
                "cost_per_test": 0.055137,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 10.0,
                "latency": 229.682,
                "stderr": 5.571,
                "cost_per_test": 0.532775,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 10.0,
                "latency": 1152.919,
                "stderr": 5.571,
                "cost_per_test": 0.18555,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 6.667,
                "latency": 19.508,
                "stderr": 4.632,
                "cost_per_test": 0.013992,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 6.667,
                "latency": 188.923,
                "stderr": 4.632,
                "cost_per_test": 0.290063,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 6.667,
                "latency": 372.619,
                "stderr": 4.632,
                "cost_per_test": 0.056492,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 6.667,
                "latency": 618.768,
                "stderr": 4.632,
                "cost_per_test": 1.063116,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 3.333,
                "latency": 51.629,
                "stderr": 3.333,
                "cost_per_test": 0.048442,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 3.333,
                "latency": 474.816,
                "stderr": 3.333,
                "cost_per_test": 0.199185,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 232.734,
                "stderr": 0.0,
                "cost_per_test": 0.036118,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.0,
                "latency": 396.297,
                "stderr": 0.0,
                "cost_per_test": 0.09495,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 0.0,
                "latency": 661.96,
                "stderr": 0.0,
                "cost_per_test": 0.005274,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "immigration": {
            "meta/muse_spark_1_2": {
                "accuracy": 54.545,
                "latency": 372.58,
                "stderr": 15.746,
                "cost_per_test": 1.011872,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 54.545,
                "latency": 1545.533,
                "stderr": 15.746,
                "cost_per_test": 4.782473,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 54.545,
                "latency": 2227.72,
                "stderr": 15.746,
                "cost_per_test": 2.139853,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 54.545,
                "latency": 7783.489,
                "stderr": 15.746,
                "cost_per_test": 28.846858,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 45.455,
                "latency": 377.453,
                "stderr": 15.746,
                "cost_per_test": 0.749015,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 45.455,
                "latency": 852.202,
                "stderr": 15.746,
                "cost_per_test": 4.505049,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 45.455,
                "latency": 1055.89,
                "stderr": 15.746,
                "cost_per_test": 4.615585,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 45.455,
                "latency": 1309.204,
                "stderr": 15.746,
                "cost_per_test": 5.041009,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 45.455,
                "latency": 2577.035,
                "stderr": 15.746,
                "cost_per_test": 7.345528,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 45.455,
                "latency": 2999.169,
                "stderr": 15.746,
                "cost_per_test": 1.324131,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 45.455,
                "latency": 3576.368,
                "stderr": 15.746,
                "cost_per_test": 0.63176,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 45.455,
                "latency": 3943.62,
                "stderr": 15.746,
                "cost_per_test": 3.006031,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 45.455,
                "latency": 4168.205,
                "stderr": 15.746,
                "cost_per_test": 0.281372,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 45.455,
                "latency": 5565.396,
                "stderr": 15.746,
                "cost_per_test": 29.782421,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 45.455,
                "latency": 6121.239,
                "stderr": 15.746,
                "cost_per_test": 15.318633,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 36.364,
                "latency": 145.322,
                "stderr": 15.212,
                "cost_per_test": 0.576237,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 36.364,
                "latency": 522.726,
                "stderr": 15.212,
                "cost_per_test": 3.815712,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 36.364,
                "latency": 771.306,
                "stderr": 15.212,
                "cost_per_test": 1.548677,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 36.364,
                "latency": 1059.33,
                "stderr": 15.212,
                "cost_per_test": 0.440884,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 36.364,
                "latency": 1086.925,
                "stderr": 15.212,
                "cost_per_test": 0.562099,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 36.364,
                "latency": 1290.179,
                "stderr": 15.212,
                "cost_per_test": 9.559148,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 36.364,
                "latency": 1629.748,
                "stderr": 15.212,
                "cost_per_test": 2.754829,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 36.364,
                "latency": 1914.142,
                "stderr": 15.212,
                "cost_per_test": 6.6645,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 36.364,
                "latency": 2079.496,
                "stderr": 15.212,
                "cost_per_test": 14.716249,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 36.364,
                "latency": 2188.906,
                "stderr": 15.212,
                "cost_per_test": 0.050144,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 36.364,
                "latency": 2309.031,
                "stderr": 15.212,
                "cost_per_test": 5.522954,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 36.364,
                "latency": 2363.885,
                "stderr": 15.212,
                "cost_per_test": 0.899309,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 36.364,
                "latency": 2512.615,
                "stderr": 15.212,
                "cost_per_test": 8.597148,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 36.364,
                "latency": 2795.76,
                "stderr": 15.212,
                "cost_per_test": 1.125625,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 36.364,
                "latency": 4229.709,
                "stderr": 15.212,
                "cost_per_test": 29.748136,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 36.364,
                "latency": 5391.649,
                "stderr": 15.212,
                "cost_per_test": 2.04155,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 27.273,
                "latency": 128.854,
                "stderr": 14.084,
                "cost_per_test": 0.088481,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 27.273,
                "latency": 176.78,
                "stderr": 14.084,
                "cost_per_test": 0.283317,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 27.273,
                "latency": 231.193,
                "stderr": 14.084,
                "cost_per_test": 1.096453,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 27.273,
                "latency": 376.011,
                "stderr": 14.084,
                "cost_per_test": 0.081151,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 27.273,
                "latency": 586.852,
                "stderr": 14.084,
                "cost_per_test": 0.921916,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 27.273,
                "latency": 659.266,
                "stderr": 14.084,
                "cost_per_test": 0.026153,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 27.273,
                "latency": 728.3,
                "stderr": 14.084,
                "cost_per_test": 3.441413,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 27.273,
                "latency": 733.81,
                "stderr": 14.084,
                "cost_per_test": 0.307991,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 27.273,
                "latency": 866.165,
                "stderr": 14.084,
                "cost_per_test": 0.178499,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 27.273,
                "latency": 882.27,
                "stderr": 14.084,
                "cost_per_test": 0.482243,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 27.273,
                "latency": 1048.83,
                "stderr": 14.084,
                "cost_per_test": 2.703862,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 27.273,
                "latency": 1192.0,
                "stderr": 14.084,
                "cost_per_test": 1.003117,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 27.273,
                "latency": 1260.914,
                "stderr": 14.084,
                "cost_per_test": 0.296663,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 27.273,
                "latency": 1537.666,
                "stderr": 14.084,
                "cost_per_test": 0.570169,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 27.273,
                "latency": 1793.408,
                "stderr": 14.084,
                "cost_per_test": 0.234002,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 27.273,
                "latency": 2067.692,
                "stderr": 14.084,
                "cost_per_test": 0.952254,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 27.273,
                "latency": 2091.212,
                "stderr": 14.084,
                "cost_per_test": 8.705325,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 27.273,
                "latency": 2203.778,
                "stderr": 14.084,
                "cost_per_test": 2.644131,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 27.273,
                "latency": 4515.312,
                "stderr": 14.084,
                "cost_per_test": 1.089,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 27.273,
                "latency": 4767.048,
                "stderr": 14.084,
                "cost_per_test": 3.223316,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 27.273,
                "latency": 5257.651,
                "stderr": 14.084,
                "cost_per_test": 3.326554,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 27.273,
                "latency": 7303.148,
                "stderr": 14.084,
                "cost_per_test": 4.997493,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 18.182,
                "latency": 60.521,
                "stderr": 12.197,
                "cost_per_test": 0.102995,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 18.182,
                "latency": 69.703,
                "stderr": 12.197,
                "cost_per_test": 0.06269,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 18.182,
                "latency": 71.817,
                "stderr": 12.197,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 18.182,
                "latency": 147.165,
                "stderr": 12.197,
                "cost_per_test": 0.405548,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 18.182,
                "latency": 185.396,
                "stderr": 12.197,
                "cost_per_test": 0.389491,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 18.182,
                "latency": 219.971,
                "stderr": 12.197,
                "cost_per_test": 0.884226,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 18.182,
                "latency": 255.303,
                "stderr": 12.197,
                "cost_per_test": 0.300956,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 18.182,
                "latency": 389.141,
                "stderr": 12.197,
                "cost_per_test": 0.560419,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 18.182,
                "latency": 449.603,
                "stderr": 12.197,
                "cost_per_test": 0.076601,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 18.182,
                "latency": 598.99,
                "stderr": 12.197,
                "cost_per_test": 0.650754,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 18.182,
                "latency": 705.891,
                "stderr": 12.197,
                "cost_per_test": 1.495632,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 18.182,
                "latency": 812.146,
                "stderr": 12.197,
                "cost_per_test": 0.150825,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 18.182,
                "latency": 823.783,
                "stderr": 12.197,
                "cost_per_test": 0.603054,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 18.182,
                "latency": 1272.83,
                "stderr": 12.197,
                "cost_per_test": 0.124704,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 18.182,
                "latency": 1561.245,
                "stderr": 12.197,
                "cost_per_test": 1.49714,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 18.182,
                "latency": 2027.42,
                "stderr": 12.197,
                "cost_per_test": 1.631269,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 18.182,
                "latency": 2445.483,
                "stderr": 12.197,
                "cost_per_test": 3.066554,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 18.182,
                "latency": 2966.232,
                "stderr": 12.197,
                "cost_per_test": 1.017586,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 18.182,
                "latency": 3025.432,
                "stderr": 12.197,
                "cost_per_test": 5.242893,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 18.182,
                "latency": 3289.995,
                "stderr": 12.197,
                "cost_per_test": 0.523583,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 18.182,
                "latency": 5146.748,
                "stderr": 12.197,
                "cost_per_test": 9.402071,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 18.182,
                "latency": 6986.35,
                "stderr": 12.197,
                "cost_per_test": 2.97227,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 18.182,
                "latency": 7116.076,
                "stderr": 12.197,
                "cost_per_test": 8.488087,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 9.091,
                "latency": 21.669,
                "stderr": 9.091,
                "cost_per_test": 0.014207,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 9.091,
                "latency": 242.659,
                "stderr": 9.091,
                "cost_per_test": 0.366775,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 9.091,
                "latency": 305.943,
                "stderr": 9.091,
                "cost_per_test": 0.448482,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 9.091,
                "latency": 473.789,
                "stderr": 9.091,
                "cost_per_test": 0.036168,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 9.091,
                "latency": 848.642,
                "stderr": 9.091,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 9.091,
                "latency": 851.042,
                "stderr": 9.091,
                "cost_per_test": 1.91176,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 9.091,
                "latency": 902.658,
                "stderr": 9.091,
                "cost_per_test": 0.700549,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 9.091,
                "latency": 5272.469,
                "stderr": 9.091,
                "cost_per_test": 3.88229,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 0.0,
                "latency": 38.568,
                "stderr": 0.0,
                "cost_per_test": 0.046126,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 0.0,
                "latency": 173.024,
                "stderr": 0.0,
                "cost_per_test": 0.003083,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 0.0,
                "latency": 201.446,
                "stderr": 0.0,
                "cost_per_test": 0.033288,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 0.0,
                "latency": 356.501,
                "stderr": 0.0,
                "cost_per_test": 0.229174,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 0.0,
                "latency": 394.752,
                "stderr": 0.0,
                "cost_per_test": 0.099916,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 0.0,
                "latency": 559.812,
                "stderr": 0.0,
                "cost_per_test": 0.159001,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 0.0,
                "latency": 610.78,
                "stderr": 0.0,
                "cost_per_test": 0.296042,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            }
        },
        "weighted_pass": {
            "google/gemini-4-carbon": {
                "accuracy": 91.609,
                "latency": 3112.579,
                "stderr": 1.026,
                "cost_per_test": 8.743524,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 90.991,
                "latency": 1881.045,
                "stderr": 1.028,
                "cost_per_test": 6.44908,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 90.577,
                "latency": 1860.719,
                "stderr": 1.151,
                "cost_per_test": 6.762498,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 90.213,
                "latency": 3449.382,
                "stderr": 1.224,
                "cost_per_test": 23.055789,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 89.804,
                "latency": 1654.378,
                "stderr": 1.212,
                "cost_per_test": 3.595401,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 88.346,
                "latency": 1504.077,
                "stderr": 1.25,
                "cost_per_test": 1.530204,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 87.872,
                "latency": 735.191,
                "stderr": 1.32,
                "cost_per_test": 0.288586,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 87.816,
                "latency": 4631.383,
                "stderr": 1.262,
                "cost_per_test": 21.605566,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 87.765,
                "latency": 1337.604,
                "stderr": 1.32,
                "cost_per_test": 9.791426,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 87.732,
                "latency": 3110.171,
                "stderr": 1.376,
                "cost_per_test": 2.241388,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 87.726,
                "latency": 927.111,
                "stderr": 1.368,
                "cost_per_test": 3.980242,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 87.541,
                "latency": 3067.486,
                "stderr": 1.273,
                "cost_per_test": 0.240392,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 87.457,
                "latency": 324.318,
                "stderr": 1.293,
                "cost_per_test": 1.630498,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 86.8,
                "latency": 317.461,
                "stderr": 1.569,
                "cost_per_test": 0.588812,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 86.551,
                "latency": 1274.337,
                "stderr": 1.374,
                "cost_per_test": 4.920812,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 86.28,
                "latency": 2106.881,
                "stderr": 1.376,
                "cost_per_test": 0.69856,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 86.146,
                "latency": 2388.741,
                "stderr": 1.44,
                "cost_per_test": 0.04878,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 86.087,
                "latency": 3677.636,
                "stderr": 1.463,
                "cost_per_test": 2.486351,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 86.067,
                "latency": 714.036,
                "stderr": 1.441,
                "cost_per_test": 2.820942,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 86.027,
                "latency": 2313.552,
                "stderr": 1.32,
                "cost_per_test": 3.674091,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 85.989,
                "latency": 1763.023,
                "stderr": 1.394,
                "cost_per_test": 1.559385,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 85.493,
                "latency": 1816.322,
                "stderr": 1.537,
                "cost_per_test": 0.181201,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 85.339,
                "latency": 989.604,
                "stderr": 1.359,
                "cost_per_test": 2.302948,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 85.271,
                "latency": 5571.099,
                "stderr": 1.387,
                "cost_per_test": 2.723732,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 85.266,
                "latency": 306.893,
                "stderr": 1.523,
                "cost_per_test": 0.508097,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 85.209,
                "latency": 530.813,
                "stderr": 1.592,
                "cost_per_test": 0.650063,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 85.176,
                "latency": 3839.511,
                "stderr": 1.486,
                "cost_per_test": 0.846372,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 85.092,
                "latency": 4172.869,
                "stderr": 1.359,
                "cost_per_test": 7.907299,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 85.047,
                "latency": 2373.859,
                "stderr": 1.285,
                "cost_per_test": 0.845008,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 84.765,
                "latency": 1491.339,
                "stderr": 1.293,
                "cost_per_test": 10.577143,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 84.764,
                "latency": 4979.806,
                "stderr": 1.505,
                "cost_per_test": 1.647359,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 84.759,
                "latency": 2168.413,
                "stderr": 1.471,
                "cost_per_test": 0.838645,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 84.643,
                "latency": 760.122,
                "stderr": 1.483,
                "cost_per_test": 3.469601,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 84.621,
                "latency": 312.994,
                "stderr": 1.419,
                "cost_per_test": 0.377379,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 84.592,
                "latency": 1545.997,
                "stderr": 1.431,
                "cost_per_test": 2.719446,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 84.522,
                "latency": 1704.14,
                "stderr": 1.421,
                "cost_per_test": 1.09052,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 84.503,
                "latency": 701.783,
                "stderr": 1.427,
                "cost_per_test": 0.24793,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 84.423,
                "latency": 953.767,
                "stderr": 1.429,
                "cost_per_test": 1.086814,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 84.312,
                "latency": 2060.48,
                "stderr": 1.321,
                "cost_per_test": 3.384561,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 84.274,
                "latency": 5588.989,
                "stderr": 1.74,
                "cost_per_test": 13.663179,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 84.132,
                "latency": 288.96,
                "stderr": 1.519,
                "cost_per_test": 0.760397,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 84.089,
                "latency": 511.955,
                "stderr": 1.475,
                "cost_per_test": 3.05609,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 83.964,
                "latency": 2265.105,
                "stderr": 1.474,
                "cost_per_test": 7.402471,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 83.93,
                "latency": 277.753,
                "stderr": 1.373,
                "cost_per_test": 1.305281,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 83.843,
                "latency": 2527.177,
                "stderr": 1.518,
                "cost_per_test": 1.088017,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 83.076,
                "latency": 2911.983,
                "stderr": 1.532,
                "cost_per_test": 0.444777,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 82.871,
                "latency": 1023.468,
                "stderr": 1.478,
                "cost_per_test": 0.889209,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 82.633,
                "latency": 1600.721,
                "stderr": 1.581,
                "cost_per_test": 2.159719,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 82.605,
                "latency": 5455.029,
                "stderr": 1.912,
                "cost_per_test": 2.222804,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 81.223,
                "latency": 1266.744,
                "stderr": 1.513,
                "cost_per_test": 0.275172,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 81.146,
                "latency": 813.838,
                "stderr": 1.532,
                "cost_per_test": 0.342955,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 80.689,
                "latency": 1024.619,
                "stderr": 1.537,
                "cost_per_test": 0.682218,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 80.669,
                "latency": 7649.584,
                "stderr": 2.232,
                "cost_per_test": 5.312295,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 80.543,
                "latency": 168.482,
                "stderr": 1.573,
                "cost_per_test": 0.466741,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 80.378,
                "latency": 6749.531,
                "stderr": 2.238,
                "cost_per_test": 23.613691,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 80.193,
                "latency": 869.377,
                "stderr": 1.522,
                "cost_per_test": 0.494044,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 80.072,
                "latency": 446.517,
                "stderr": 1.599,
                "cost_per_test": 0.074351,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 79.577,
                "latency": 919.689,
                "stderr": 1.844,
                "cost_per_test": 0.070424,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 79.467,
                "latency": 94.568,
                "stderr": 1.578,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 79.296,
                "latency": 2547.747,
                "stderr": 1.631,
                "cost_per_test": 0.440296,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 79.181,
                "latency": 1449.451,
                "stderr": 1.564,
                "cost_per_test": 4.897717,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 78.924,
                "latency": 774.823,
                "stderr": 1.605,
                "cost_per_test": 0.141548,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 78.708,
                "latency": 4202.728,
                "stderr": 1.61,
                "cost_per_test": 3.220614,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 78.4,
                "latency": 808.802,
                "stderr": 1.689,
                "cost_per_test": 1.400701,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 77.808,
                "latency": 325.824,
                "stderr": 1.696,
                "cost_per_test": 0.385737,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 76.61,
                "latency": 242.05,
                "stderr": 1.602,
                "cost_per_test": 1.100145,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 76.0,
                "latency": 416.437,
                "stderr": 1.753,
                "cost_per_test": 0.082648,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 75.174,
                "latency": 187.7,
                "stderr": 1.651,
                "cost_per_test": 0.135395,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 74.98,
                "latency": 831.995,
                "stderr": 1.746,
                "cost_per_test": 0.295949,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 74.742,
                "latency": 1853.133,
                "stderr": 1.698,
                "cost_per_test": 1.520305,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 74.341,
                "latency": 658.954,
                "stderr": 1.708,
                "cost_per_test": 0.697997,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 73.21,
                "latency": 515.464,
                "stderr": 1.797,
                "cost_per_test": 0.028227,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 71.447,
                "latency": 6237.394,
                "stderr": 2.566,
                "cost_per_test": 7.231463,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 71.336,
                "latency": 1332.514,
                "stderr": 1.765,
                "cost_per_test": 0.23035,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 70.148,
                "latency": 82.971,
                "stderr": 1.855,
                "cost_per_test": 0.063761,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 69.491,
                "latency": 84.583,
                "stderr": 1.827,
                "cost_per_test": 0.203143,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 68.939,
                "latency": 59.197,
                "stderr": 1.846,
                "cost_per_test": 0.124723,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 68.216,
                "latency": 391.469,
                "stderr": 1.898,
                "cost_per_test": 0.033318,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 68.005,
                "latency": 611.366,
                "stderr": 1.98,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 67.326,
                "latency": 2256.339,
                "stderr": 1.837,
                "cost_per_test": 1.864673,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 65.11,
                "latency": 244.487,
                "stderr": 1.904,
                "cost_per_test": 0.261499,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 64.939,
                "latency": 243.38,
                "stderr": 1.946,
                "cost_per_test": 0.26891,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 64.627,
                "latency": 211.833,
                "stderr": 1.989,
                "cost_per_test": 0.036023,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 63.885,
                "latency": 420.647,
                "stderr": 1.987,
                "cost_per_test": 0.219167,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 63.308,
                "latency": 613.248,
                "stderr": 1.96,
                "cost_per_test": 1.308079,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 62.338,
                "latency": 466.26,
                "stderr": 1.84,
                "cost_per_test": 0.219839,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 61.958,
                "latency": 526.838,
                "stderr": 1.95,
                "cost_per_test": 0.141253,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 58.67,
                "latency": 47.609,
                "stderr": 1.894,
                "cost_per_test": 0.058112,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 56.27,
                "latency": 352.036,
                "stderr": 1.989,
                "cost_per_test": 0.107892,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 54.415,
                "latency": 20.403,
                "stderr": 1.956,
                "cost_per_test": 0.013607,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 48.972,
                "latency": 285.48,
                "stderr": 1.864,
                "cost_per_test": 0.004365,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__administrative_regulatory": {
            "google/gemini-4-argon": {
                "accuracy": 92.945,
                "latency": 1486.33,
                "stderr": 1.862,
                "cost_per_test": 4.807402,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 92.809,
                "latency": 1691.752,
                "stderr": 1.778,
                "cost_per_test": 3.18787,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 92.676,
                "latency": 1532.62,
                "stderr": 1.919,
                "cost_per_test": 6.583578,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 92.674,
                "latency": 631.256,
                "stderr": 1.865,
                "cost_per_test": 0.243078,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 92.52,
                "latency": 2535.332,
                "stderr": 2.03,
                "cost_per_test": 6.803146,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 92.519,
                "latency": 3106.859,
                "stderr": 2.145,
                "cost_per_test": 20.305794,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 91.39,
                "latency": 1372.59,
                "stderr": 2.119,
                "cost_per_test": 1.517644,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 91.346,
                "latency": 4690.9,
                "stderr": 2.288,
                "cost_per_test": 11.667322,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 91.277,
                "latency": 1931.905,
                "stderr": 1.979,
                "cost_per_test": 0.652538,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 91.07,
                "latency": 2956.379,
                "stderr": 2.219,
                "cost_per_test": 2.133253,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 90.94,
                "latency": 2038.532,
                "stderr": 2.085,
                "cost_per_test": 3.174554,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 90.822,
                "latency": 2848.785,
                "stderr": 1.911,
                "cost_per_test": 0.198792,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 90.69,
                "latency": 4138.255,
                "stderr": 2.118,
                "cost_per_test": 19.689902,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 90.571,
                "latency": 889.246,
                "stderr": 2.422,
                "cost_per_test": 3.922674,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 90.562,
                "latency": 1523.131,
                "stderr": 2.0,
                "cost_per_test": 0.921373,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 90.519,
                "latency": 3192.985,
                "stderr": 2.201,
                "cost_per_test": 2.201925,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 90.224,
                "latency": 1179.378,
                "stderr": 2.232,
                "cost_per_test": 9.381311,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 90.108,
                "latency": 2358.521,
                "stderr": 1.981,
                "cost_per_test": 0.049642,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 89.252,
                "latency": 749.583,
                "stderr": 2.267,
                "cost_per_test": 3.444134,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 89.06,
                "latency": 656.731,
                "stderr": 2.16,
                "cost_per_test": 0.246492,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 88.759,
                "latency": 3607.895,
                "stderr": 2.29,
                "cost_per_test": 6.866314,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 88.75,
                "latency": 1001.46,
                "stderr": 2.226,
                "cost_per_test": 1.36196,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 88.676,
                "latency": 3941.13,
                "stderr": 2.238,
                "cost_per_test": 1.441703,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 88.497,
                "latency": 1643.295,
                "stderr": 2.31,
                "cost_per_test": 2.2119,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 88.492,
                "latency": 5582.268,
                "stderr": 2.189,
                "cost_per_test": 2.616141,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 88.325,
                "latency": 3713.178,
                "stderr": 2.506,
                "cost_per_test": 0.834838,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 88.147,
                "latency": 302.672,
                "stderr": 2.538,
                "cost_per_test": 0.506511,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 87.942,
                "latency": 1671.25,
                "stderr": 2.515,
                "cost_per_test": 1.423111,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 87.81,
                "latency": 1285.722,
                "stderr": 2.292,
                "cost_per_test": 2.507811,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 87.76,
                "latency": 2171.348,
                "stderr": 2.3,
                "cost_per_test": 0.862352,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 87.613,
                "latency": 296.369,
                "stderr": 2.448,
                "cost_per_test": 1.661721,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 87.513,
                "latency": 2245.491,
                "stderr": 2.206,
                "cost_per_test": 0.755621,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 87.482,
                "latency": 1103.075,
                "stderr": 2.459,
                "cost_per_test": 8.442858,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 87.402,
                "latency": 1609.837,
                "stderr": 2.774,
                "cost_per_test": 0.17954,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 87.365,
                "latency": 636.543,
                "stderr": 2.257,
                "cost_per_test": 2.731524,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 87.214,
                "latency": 1016.365,
                "stderr": 2.694,
                "cost_per_test": 4.175009,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 87.126,
                "latency": 285.017,
                "stderr": 2.45,
                "cost_per_test": 1.356228,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 87.035,
                "latency": 7602.624,
                "stderr": 3.249,
                "cost_per_test": 5.772668,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 87.011,
                "latency": 425.156,
                "stderr": 2.677,
                "cost_per_test": 3.06182,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 86.797,
                "latency": 901.446,
                "stderr": 2.273,
                "cost_per_test": 2.410326,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 86.761,
                "latency": 306.876,
                "stderr": 3.007,
                "cost_per_test": 0.540035,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 86.669,
                "latency": 2093.383,
                "stderr": 2.717,
                "cost_per_test": 7.953883,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 86.33,
                "latency": 2424.284,
                "stderr": 2.663,
                "cost_per_test": 1.073701,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 86.153,
                "latency": 755.036,
                "stderr": 2.391,
                "cost_per_test": 0.352926,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 85.95,
                "latency": 836.031,
                "stderr": 2.455,
                "cost_per_test": 0.808378,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 85.884,
                "latency": 481.922,
                "stderr": 3.022,
                "cost_per_test": 0.624599,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 85.547,
                "latency": 1439.965,
                "stderr": 2.599,
                "cost_per_test": 2.756725,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 85.427,
                "latency": 5979.926,
                "stderr": 3.718,
                "cost_per_test": 20.862676,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 85.147,
                "latency": 244.664,
                "stderr": 2.799,
                "cost_per_test": 0.704178,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 85.049,
                "latency": 2966.067,
                "stderr": 2.723,
                "cost_per_test": 0.472211,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 84.8,
                "latency": 3381.142,
                "stderr": 2.625,
                "cost_per_test": 2.810214,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 84.144,
                "latency": 877.92,
                "stderr": 2.374,
                "cost_per_test": 0.44143,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 83.971,
                "latency": 789.443,
                "stderr": 2.671,
                "cost_per_test": 1.359392,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 83.878,
                "latency": 1185.766,
                "stderr": 2.437,
                "cost_per_test": 0.295871,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 83.623,
                "latency": 282.392,
                "stderr": 2.846,
                "cost_per_test": 0.382532,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 83.517,
                "latency": 139.396,
                "stderr": 2.559,
                "cost_per_test": 0.409841,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 83.372,
                "latency": 903.624,
                "stderr": 2.545,
                "cost_per_test": 0.671217,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 83.349,
                "latency": 715.742,
                "stderr": 3.224,
                "cost_per_test": 0.066967,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 83.175,
                "latency": 2341.023,
                "stderr": 2.64,
                "cost_per_test": 0.409481,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 83.099,
                "latency": 94.03,
                "stderr": 2.713,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 82.955,
                "latency": 842.128,
                "stderr": 2.398,
                "cost_per_test": 0.134944,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 82.011,
                "latency": 5301.951,
                "stderr": 4.0,
                "cost_per_test": 2.211937,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 81.914,
                "latency": 5673.421,
                "stderr": 3.783,
                "cost_per_test": 6.338985,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 81.478,
                "latency": 434.539,
                "stderr": 2.787,
                "cost_per_test": 0.075638,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 81.313,
                "latency": 166.672,
                "stderr": 2.681,
                "cost_per_test": 0.129186,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 80.457,
                "latency": 1112.655,
                "stderr": 2.933,
                "cost_per_test": 3.887103,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 80.005,
                "latency": 307.232,
                "stderr": 3.081,
                "cost_per_test": 0.323431,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 79.948,
                "latency": 805.914,
                "stderr": 2.796,
                "cost_per_test": 0.359357,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 79.215,
                "latency": 219.629,
                "stderr": 2.633,
                "cost_per_test": 1.225685,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 77.055,
                "latency": 381.12,
                "stderr": 3.191,
                "cost_per_test": 0.087857,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 76.623,
                "latency": 625.818,
                "stderr": 2.914,
                "cost_per_test": 0.670062,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 75.531,
                "latency": 1689.904,
                "stderr": 3.082,
                "cost_per_test": 1.488125,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 75.15,
                "latency": 494.675,
                "stderr": 3.183,
                "cost_per_test": 0.027887,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 73.866,
                "latency": 72.945,
                "stderr": 3.085,
                "cost_per_test": 0.19233,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 72.761,
                "latency": 1153.879,
                "stderr": 3.058,
                "cost_per_test": 0.195218,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 72.398,
                "latency": 476.589,
                "stderr": 2.881,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 71.96,
                "latency": 50.778,
                "stderr": 3.222,
                "cost_per_test": 0.123387,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 71.929,
                "latency": 343.766,
                "stderr": 3.121,
                "cost_per_test": 0.038807,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 70.856,
                "latency": 69.443,
                "stderr": 3.094,
                "cost_per_test": 0.05243,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 69.72,
                "latency": 1961.008,
                "stderr": 3.312,
                "cost_per_test": 1.456649,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 67.862,
                "latency": 199.066,
                "stderr": 3.241,
                "cost_per_test": 0.274657,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 67.477,
                "latency": 382.835,
                "stderr": 3.415,
                "cost_per_test": 0.200329,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 67.472,
                "latency": 575.292,
                "stderr": 3.109,
                "cost_per_test": 1.130658,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 67.15,
                "latency": 215.774,
                "stderr": 3.505,
                "cost_per_test": 0.032184,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 66.556,
                "latency": 383.823,
                "stderr": 2.803,
                "cost_per_test": 0.163947,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 65.678,
                "latency": 402.197,
                "stderr": 3.408,
                "cost_per_test": 0.125736,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 65.064,
                "latency": 226.406,
                "stderr": 3.195,
                "cost_per_test": 0.379263,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 58.074,
                "latency": 332.137,
                "stderr": 3.205,
                "cost_per_test": 0.087793,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 55.228,
                "latency": 41.907,
                "stderr": 3.324,
                "cost_per_test": 0.046871,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 52.826,
                "latency": 18.892,
                "stderr": 3.448,
                "cost_per_test": 0.013517,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 48.382,
                "latency": 409.746,
                "stderr": 3.018,
                "cost_per_test": 0.004424,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__business_commercial": {
            "anthropic/claude-fable-5-1": {
                "accuracy": 94.617,
                "latency": 3855.165,
                "stderr": 1.435,
                "cost_per_test": 27.230592,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 92.246,
                "latency": 2163.785,
                "stderr": 2.569,
                "cost_per_test": 7.357666,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 91.87,
                "latency": 2252.454,
                "stderr": 1.752,
                "cost_per_test": 7.818685,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 91.526,
                "latency": 755.219,
                "stderr": 1.921,
                "cost_per_test": 0.250157,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 91.374,
                "latency": 1482.098,
                "stderr": 1.907,
                "cost_per_test": 9.515091,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 91.27,
                "latency": 3410.204,
                "stderr": 1.939,
                "cost_per_test": 0.266594,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 91.066,
                "latency": 315.567,
                "stderr": 1.922,
                "cost_per_test": 1.733887,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 90.979,
                "latency": 3764.53,
                "stderr": 2.033,
                "cost_per_test": 10.672138,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 89.811,
                "latency": 1748.44,
                "stderr": 3.186,
                "cost_per_test": 3.900906,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 89.75,
                "latency": 1612.05,
                "stderr": 2.478,
                "cost_per_test": 1.54459,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 89.525,
                "latency": 298.916,
                "stderr": 2.803,
                "cost_per_test": 0.773586,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 89.217,
                "latency": 328.596,
                "stderr": 2.434,
                "cost_per_test": 0.506176,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 89.091,
                "latency": 319.151,
                "stderr": 2.687,
                "cost_per_test": 0.363621,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 89.077,
                "latency": 2344.543,
                "stderr": 2.243,
                "cost_per_test": 0.811263,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 89.064,
                "latency": 3261.478,
                "stderr": 2.74,
                "cost_per_test": 2.29039,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 89.003,
                "latency": 4581.547,
                "stderr": 2.864,
                "cost_per_test": 19.023356,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 88.955,
                "latency": 1523.334,
                "stderr": 2.648,
                "cost_per_test": 1.137443,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 88.674,
                "latency": 353.92,
                "stderr": 3.166,
                "cost_per_test": 0.626202,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 88.343,
                "latency": 626.261,
                "stderr": 2.835,
                "cost_per_test": 3.528557,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 88.24,
                "latency": 581.128,
                "stderr": 2.861,
                "cost_per_test": 0.642459,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 87.972,
                "latency": 1023.821,
                "stderr": 2.977,
                "cost_per_test": 2.037941,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 87.785,
                "latency": 2247.75,
                "stderr": 2.946,
                "cost_per_test": 0.705293,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 87.673,
                "latency": 4069.494,
                "stderr": 2.959,
                "cost_per_test": 0.863275,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 87.618,
                "latency": 2633.008,
                "stderr": 2.579,
                "cost_per_test": 0.372136,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 87.601,
                "latency": 2895.709,
                "stderr": 2.247,
                "cost_per_test": 4.184362,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 87.466,
                "latency": 1106.364,
                "stderr": 2.623,
                "cost_per_test": 0.920113,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 87.402,
                "latency": 1380.854,
                "stderr": 2.776,
                "cost_per_test": 4.555398,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 87.309,
                "latency": 2334.81,
                "stderr": 2.242,
                "cost_per_test": 0.78052,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 87.024,
                "latency": 5681.673,
                "stderr": 3.122,
                "cost_per_test": 1.820007,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 86.88,
                "latency": 904.491,
                "stderr": 3.443,
                "cost_per_test": 3.808111,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 86.805,
                "latency": 913.577,
                "stderr": 2.755,
                "cost_per_test": 0.893714,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 86.764,
                "latency": 2266.219,
                "stderr": 2.947,
                "cost_per_test": 6.684872,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 86.759,
                "latency": 1790.227,
                "stderr": 2.83,
                "cost_per_test": 3.031928,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 86.756,
                "latency": 1723.973,
                "stderr": 2.963,
                "cost_per_test": 2.311817,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 86.675,
                "latency": 694.588,
                "stderr": 2.424,
                "cost_per_test": 0.226487,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 86.655,
                "latency": 3920.461,
                "stderr": 3.369,
                "cost_per_test": 2.547702,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 86.491,
                "latency": 917.178,
                "stderr": 2.924,
                "cost_per_test": 0.622362,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 86.246,
                "latency": 1961.481,
                "stderr": 2.635,
                "cost_per_test": 1.153979,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 86.128,
                "latency": 4084.965,
                "stderr": 2.961,
                "cost_per_test": 7.302701,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 86.112,
                "latency": 777.706,
                "stderr": 3.565,
                "cost_per_test": 2.888428,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 86.015,
                "latency": 5387.125,
                "stderr": 3.017,
                "cost_per_test": 1.920252,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 85.024,
                "latency": 1783.688,
                "stderr": 3.677,
                "cost_per_test": 0.170401,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 84.578,
                "latency": 173.23,
                "stderr": 3.01,
                "cost_per_test": 0.487431,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 84.545,
                "latency": 2576.157,
                "stderr": 3.631,
                "cost_per_test": 0.05022,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 84.159,
                "latency": 2357.851,
                "stderr": 3.13,
                "cost_per_test": 3.455558,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 84.143,
                "latency": 317.652,
                "stderr": 2.921,
                "cost_per_test": 1.232191,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 84.011,
                "latency": 2499.931,
                "stderr": 3.767,
                "cost_per_test": 1.025341,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 83.821,
                "latency": 1735.645,
                "stderr": 3.003,
                "cost_per_test": 10.624987,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 83.547,
                "latency": 790.908,
                "stderr": 3.543,
                "cost_per_test": 3.239177,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 83.293,
                "latency": 1087.928,
                "stderr": 3.436,
                "cost_per_test": 0.215579,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 82.74,
                "latency": 5721.537,
                "stderr": 4.469,
                "cost_per_test": 2.262179,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 82.694,
                "latency": 867.939,
                "stderr": 3.326,
                "cost_per_test": 0.34538,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 82.557,
                "latency": 498.212,
                "stderr": 2.852,
                "cost_per_test": 0.086143,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 81.655,
                "latency": 336.248,
                "stderr": 2.98,
                "cost_per_test": 0.399162,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 81.634,
                "latency": 909.412,
                "stderr": 3.932,
                "cost_per_test": 0.066442,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 81.54,
                "latency": 663.553,
                "stderr": 2.354,
                "cost_per_test": 0.73626,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 81.111,
                "latency": 452.214,
                "stderr": 3.408,
                "cost_per_test": 0.075902,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 80.966,
                "latency": 913.289,
                "stderr": 3.37,
                "cost_per_test": 1.493846,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 80.773,
                "latency": 216.623,
                "stderr": 3.085,
                "cost_per_test": 0.45295,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 80.401,
                "latency": 1674.291,
                "stderr": 3.149,
                "cost_per_test": 5.299121,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 79.089,
                "latency": 792.476,
                "stderr": 3.455,
                "cost_per_test": 0.529635,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 79.086,
                "latency": 2678.679,
                "stderr": 3.272,
                "cost_per_test": 0.455548,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 78.436,
                "latency": 1873.65,
                "stderr": 3.571,
                "cost_per_test": 1.432344,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 78.197,
                "latency": 709.804,
                "stderr": 4.24,
                "cost_per_test": 0.132473,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 78.141,
                "latency": 108.067,
                "stderr": 3.702,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 77.77,
                "latency": 6273.67,
                "stderr": 5.246,
                "cost_per_test": 15.137463,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 76.138,
                "latency": 787.349,
                "stderr": 3.499,
                "cost_per_test": 0.216443,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 75.896,
                "latency": 5207.222,
                "stderr": 3.665,
                "cost_per_test": 3.545917,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 75.874,
                "latency": 200.727,
                "stderr": 3.702,
                "cost_per_test": 0.135638,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 75.839,
                "latency": 405.023,
                "stderr": 3.853,
                "cost_per_test": 0.029283,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 75.476,
                "latency": 73.592,
                "stderr": 3.45,
                "cost_per_test": 0.170453,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 74.083,
                "latency": 489.906,
                "stderr": 3.402,
                "cost_per_test": 0.032371,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 73.216,
                "latency": 62.398,
                "stderr": 3.525,
                "cost_per_test": 0.120286,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 73.068,
                "latency": 8276.057,
                "stderr": 6.186,
                "cost_per_test": 5.198048,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 71.304,
                "latency": 637.024,
                "stderr": 3.21,
                "cost_per_test": 1.423124,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 71.172,
                "latency": 2506.739,
                "stderr": 3.437,
                "cost_per_test": 2.128528,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 71.001,
                "latency": 108.913,
                "stderr": 4.146,
                "cost_per_test": 0.08423,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 70.961,
                "latency": 7461.121,
                "stderr": 6.321,
                "cost_per_test": 26.294195,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 70.813,
                "latency": 1493.936,
                "stderr": 3.808,
                "cost_per_test": 0.245468,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 70.541,
                "latency": 620.168,
                "stderr": 4.223,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 69.927,
                "latency": 261.305,
                "stderr": 4.057,
                "cost_per_test": 0.042315,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 69.871,
                "latency": 241.456,
                "stderr": 4.105,
                "cost_per_test": 0.168225,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 68.944,
                "latency": 279.601,
                "stderr": 4.08,
                "cost_per_test": 0.268062,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 68.55,
                "latency": 343.417,
                "stderr": 4.162,
                "cost_per_test": 0.211419,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 66.395,
                "latency": 568.266,
                "stderr": 4.306,
                "cost_per_test": 0.150807,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 66.266,
                "latency": 6424.655,
                "stderr": 6.25,
                "cost_per_test": 7.199044,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 65.381,
                "latency": 44.657,
                "stderr": 3.486,
                "cost_per_test": 0.057037,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 62.671,
                "latency": 402.376,
                "stderr": 4.228,
                "cost_per_test": 0.204493,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 59.692,
                "latency": 399.428,
                "stderr": 3.772,
                "cost_per_test": 0.12794,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 54.626,
                "latency": 23.436,
                "stderr": 3.647,
                "cost_per_test": 0.014473,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 53.678,
                "latency": 124.79,
                "stderr": 3.927,
                "cost_per_test": 0.004634,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__civil_litigation": {
            "google/gemini-4-carbon": {
                "accuracy": 96.504,
                "latency": 3262.96,
                "stderr": 0.992,
                "cost_per_test": 9.405202,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 95.082,
                "latency": 1965.433,
                "stderr": 1.26,
                "cost_per_test": 6.770674,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 94.915,
                "latency": 3586.008,
                "stderr": 2.241,
                "cost_per_test": 25.254544,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 92.925,
                "latency": 1439.733,
                "stderr": 2.59,
                "cost_per_test": 3.415131,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 92.791,
                "latency": 2079.281,
                "stderr": 2.492,
                "cost_per_test": 7.304788,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 91.539,
                "latency": 254.44,
                "stderr": 2.326,
                "cost_per_test": 1.35934,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 91.451,
                "latency": 1489.886,
                "stderr": 2.697,
                "cost_per_test": 10.123039,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 91.189,
                "latency": 557.173,
                "stderr": 3.626,
                "cost_per_test": 0.757486,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 91.057,
                "latency": 1539.561,
                "stderr": 2.905,
                "cost_per_test": 1.681204,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 90.891,
                "latency": 305.909,
                "stderr": 3.793,
                "cost_per_test": 0.581163,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 90.874,
                "latency": 2608.688,
                "stderr": 2.47,
                "cost_per_test": 0.055323,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 90.836,
                "latency": 3900.655,
                "stderr": 2.889,
                "cost_per_test": 0.854674,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 90.753,
                "latency": 806.663,
                "stderr": 2.985,
                "cost_per_test": 2.997652,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 90.695,
                "latency": 2417.44,
                "stderr": 3.073,
                "cost_per_test": 0.338953,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 90.225,
                "latency": 3816.903,
                "stderr": 2.781,
                "cost_per_test": 18.560919,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 89.997,
                "latency": 1439.49,
                "stderr": 2.704,
                "cost_per_test": 5.606604,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 89.995,
                "latency": 2136.958,
                "stderr": 2.765,
                "cost_per_test": 0.917209,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 89.911,
                "latency": 2589.489,
                "stderr": 3.291,
                "cost_per_test": 2.051403,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 89.586,
                "latency": 3261.102,
                "stderr": 3.159,
                "cost_per_test": 0.259741,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 89.309,
                "latency": 5187.901,
                "stderr": 3.946,
                "cost_per_test": 2.080453,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 89.018,
                "latency": 885.474,
                "stderr": 3.3,
                "cost_per_test": 4.274558,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 88.953,
                "latency": 308.847,
                "stderr": 3.252,
                "cost_per_test": 0.367043,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 88.914,
                "latency": 288.591,
                "stderr": 2.93,
                "cost_per_test": 0.719234,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 88.748,
                "latency": 1225.988,
                "stderr": 2.752,
                "cost_per_test": 1.215179,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 88.718,
                "latency": 1956.425,
                "stderr": 2.99,
                "cost_per_test": 3.400541,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 88.711,
                "latency": 1104.503,
                "stderr": 2.935,
                "cost_per_test": 0.246003,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 88.669,
                "latency": 1743.79,
                "stderr": 2.508,
                "cost_per_test": 13.735474,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 88.446,
                "latency": 1554.157,
                "stderr": 2.957,
                "cost_per_test": 1.138521,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 88.404,
                "latency": 744.696,
                "stderr": 4.311,
                "cost_per_test": 0.267966,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 88.253,
                "latency": 1273.914,
                "stderr": 3.238,
                "cost_per_test": 1.590079,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 88.182,
                "latency": 538.692,
                "stderr": 3.314,
                "cost_per_test": 2.902212,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 87.708,
                "latency": 2447.426,
                "stderr": 2.161,
                "cost_per_test": 3.801126,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 87.667,
                "latency": 2240.6,
                "stderr": 3.339,
                "cost_per_test": 0.207888,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 87.616,
                "latency": 2408.394,
                "stderr": 3.561,
                "cost_per_test": 7.00366,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 87.539,
                "latency": 3932.081,
                "stderr": 3.014,
                "cost_per_test": 7.563297,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 87.434,
                "latency": 1001.064,
                "stderr": 2.883,
                "cost_per_test": 2.550366,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 87.42,
                "latency": 304.101,
                "stderr": 3.571,
                "cost_per_test": 0.602276,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 87.147,
                "latency": 2689.198,
                "stderr": 3.087,
                "cost_per_test": 4.546668,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 87.099,
                "latency": 663.732,
                "stderr": 3.706,
                "cost_per_test": 0.239442,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 86.955,
                "latency": 885.063,
                "stderr": 3.446,
                "cost_per_test": 0.857222,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 86.308,
                "latency": 255.111,
                "stderr": 2.832,
                "cost_per_test": 1.230964,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 85.96,
                "latency": 1095.538,
                "stderr": 3.534,
                "cost_per_test": 0.736017,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 85.824,
                "latency": 5842.182,
                "stderr": 3.368,
                "cost_per_test": 2.843836,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 85.785,
                "latency": 2280.787,
                "stderr": 3.712,
                "cost_per_test": 1.099633,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 85.331,
                "latency": 192.418,
                "stderr": 2.941,
                "cost_per_test": 0.477549,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 84.98,
                "latency": 740.554,
                "stderr": 3.193,
                "cost_per_test": 0.559142,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 84.854,
                "latency": 823.116,
                "stderr": 2.956,
                "cost_per_test": 0.385069,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 84.852,
                "latency": 2088.431,
                "stderr": 4.591,
                "cost_per_test": 0.658074,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 84.826,
                "latency": 2149.371,
                "stderr": 3.355,
                "cost_per_test": 0.78097,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 84.638,
                "latency": 493.563,
                "stderr": 3.616,
                "cost_per_test": 0.076555,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 84.566,
                "latency": 876.732,
                "stderr": 3.007,
                "cost_per_test": 1.504309,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 84.186,
                "latency": 3478.586,
                "stderr": 4.031,
                "cost_per_test": 2.299147,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 83.776,
                "latency": 838.559,
                "stderr": 3.96,
                "cost_per_test": 4.229889,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 83.513,
                "latency": 361.704,
                "stderr": 3.846,
                "cost_per_test": 0.473442,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 83.098,
                "latency": 1304.611,
                "stderr": 3.939,
                "cost_per_test": 1.737788,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 83.039,
                "latency": 5529.804,
                "stderr": 5.588,
                "cost_per_test": 1.750979,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 82.829,
                "latency": 840.078,
                "stderr": 4.722,
                "cost_per_test": 0.059764,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 82.707,
                "latency": 225.619,
                "stderr": 2.966,
                "cost_per_test": 0.508937,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 81.857,
                "latency": 92.542,
                "stderr": 2.907,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 81.699,
                "latency": 6231.017,
                "stderr": 6.129,
                "cost_per_test": 15.071116,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 81.523,
                "latency": 676.765,
                "stderr": 4.293,
                "cost_per_test": 0.13253,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 81.331,
                "latency": 775.441,
                "stderr": 3.74,
                "cost_per_test": 0.228401,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 81.028,
                "latency": 7758.778,
                "stderr": 6.771,
                "cost_per_test": 5.531551,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 80.907,
                "latency": 486.481,
                "stderr": 4.255,
                "cost_per_test": 0.080617,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 80.55,
                "latency": 4095.85,
                "stderr": 2.928,
                "cost_per_test": 3.018691,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 80.484,
                "latency": 478.114,
                "stderr": 3.245,
                "cost_per_test": 0.031614,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 78.633,
                "latency": 2593.828,
                "stderr": 4.852,
                "cost_per_test": 0.46119,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 78.225,
                "latency": 89.721,
                "stderr": 5.121,
                "cost_per_test": 0.06691,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 77.964,
                "latency": 1441.302,
                "stderr": 4.184,
                "cost_per_test": 5.211283,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 77.766,
                "latency": 246.451,
                "stderr": 3.905,
                "cost_per_test": 0.200307,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 77.536,
                "latency": 698.535,
                "stderr": 5.049,
                "cost_per_test": 0.632609,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 77.196,
                "latency": 1372.833,
                "stderr": 4.487,
                "cost_per_test": 0.220822,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 77.014,
                "latency": 215.912,
                "stderr": 4.021,
                "cost_per_test": 0.132078,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 76.085,
                "latency": 186.413,
                "stderr": 4.453,
                "cost_per_test": 0.040156,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 75.83,
                "latency": 6255.777,
                "stderr": 6.713,
                "cost_per_test": 7.220094,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 75.8,
                "latency": 1733.848,
                "stderr": 4.147,
                "cost_per_test": 1.440762,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 74.02,
                "latency": 6638.643,
                "stderr": 7.86,
                "cost_per_test": 22.569939,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 71.791,
                "latency": 428.141,
                "stderr": 5.165,
                "cost_per_test": 0.242453,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 71.755,
                "latency": 416.27,
                "stderr": 5.091,
                "cost_per_test": 0.037073,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 71.52,
                "latency": 690.614,
                "stderr": 5.401,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 70.628,
                "latency": 277.344,
                "stderr": 4.672,
                "cost_per_test": 0.287416,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 68.852,
                "latency": 68.656,
                "stderr": 4.95,
                "cost_per_test": 0.19461,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 68.638,
                "latency": 725.151,
                "stderr": 4.518,
                "cost_per_test": 0.180106,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 68.21,
                "latency": 56.454,
                "stderr": 5.443,
                "cost_per_test": 0.108873,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 67.267,
                "latency": 2120.497,
                "stderr": 5.492,
                "cost_per_test": 1.875076,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 67.094,
                "latency": 450.642,
                "stderr": 4.204,
                "cost_per_test": 0.229555,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 65.952,
                "latency": 595.409,
                "stderr": 5.685,
                "cost_per_test": 1.316462,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 65.55,
                "latency": 47.033,
                "stderr": 5.36,
                "cost_per_test": 0.064712,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 58.849,
                "latency": 19.793,
                "stderr": 5.219,
                "cost_per_test": 0.013273,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 58.132,
                "latency": 332.669,
                "stderr": 5.923,
                "cost_per_test": 0.119391,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 48.511,
                "latency": 273.794,
                "stderr": 5.134,
                "cost_per_test": 0.004442,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__constitutional_civil_rights": {
            "google/gemini-4-carbon": {
                "accuracy": 91.233,
                "latency": 2436.37,
                "stderr": 2.13,
                "cost_per_test": 6.991706,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 90.702,
                "latency": 1151.764,
                "stderr": 2.027,
                "cost_per_test": 2.721277,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 90.334,
                "latency": 1361.337,
                "stderr": 2.285,
                "cost_per_test": 4.353414,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 89.868,
                "latency": 3510.897,
                "stderr": 2.092,
                "cost_per_test": 21.64278,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 89.259,
                "latency": 632.775,
                "stderr": 2.032,
                "cost_per_test": 2.959097,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 89.18,
                "latency": 1835.592,
                "stderr": 2.515,
                "cost_per_test": 0.898996,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 88.98,
                "latency": 2583.919,
                "stderr": 2.787,
                "cost_per_test": 1.779991,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 88.843,
                "latency": 2446.246,
                "stderr": 2.289,
                "cost_per_test": 0.179145,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 88.662,
                "latency": 645.429,
                "stderr": 2.631,
                "cost_per_test": 0.225678,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 88.587,
                "latency": 1898.737,
                "stderr": 2.721,
                "cost_per_test": 6.396226,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 88.31,
                "latency": 3868.674,
                "stderr": 1.827,
                "cost_per_test": 14.526185,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 88.222,
                "latency": 1396.958,
                "stderr": 2.316,
                "cost_per_test": 1.277781,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 88.144,
                "latency": 220.844,
                "stderr": 2.458,
                "cost_per_test": 1.043129,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 88.07,
                "latency": 4751.209,
                "stderr": 2.389,
                "cost_per_test": 2.624566,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 87.91,
                "latency": 977.356,
                "stderr": 2.549,
                "cost_per_test": 6.898245,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 87.871,
                "latency": 853.113,
                "stderr": 2.478,
                "cost_per_test": 1.965994,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 87.634,
                "latency": 2063.284,
                "stderr": 2.771,
                "cost_per_test": 0.037714,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 87.039,
                "latency": 1566.47,
                "stderr": 2.548,
                "cost_per_test": 2.477637,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 87.016,
                "latency": 609.116,
                "stderr": 2.634,
                "cost_per_test": 2.462184,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 86.909,
                "latency": 1461.512,
                "stderr": 2.637,
                "cost_per_test": 0.943808,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 86.57,
                "latency": 1675.051,
                "stderr": 2.54,
                "cost_per_test": 0.53415,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 86.503,
                "latency": 1748.11,
                "stderr": 2.43,
                "cost_per_test": 2.722205,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 86.439,
                "latency": 5827.977,
                "stderr": 2.943,
                "cost_per_test": 14.117789,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 86.153,
                "latency": 1718.661,
                "stderr": 2.327,
                "cost_per_test": 0.583776,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 86.141,
                "latency": 1049.066,
                "stderr": 2.978,
                "cost_per_test": 4.00044,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 86.095,
                "latency": 558.667,
                "stderr": 2.502,
                "cost_per_test": 0.395873,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 85.897,
                "latency": 907.475,
                "stderr": 2.679,
                "cost_per_test": 0.836603,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 85.749,
                "latency": 1785.016,
                "stderr": 2.782,
                "cost_per_test": 0.721618,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 85.744,
                "latency": 1096.62,
                "stderr": 2.309,
                "cost_per_test": 0.614384,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 85.714,
                "latency": 518.106,
                "stderr": 2.752,
                "cost_per_test": 0.191689,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 85.619,
                "latency": 310.439,
                "stderr": 2.661,
                "cost_per_test": 0.303515,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 85.223,
                "latency": 224.844,
                "stderr": 2.655,
                "cost_per_test": 0.987781,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 85.201,
                "latency": 3263.997,
                "stderr": 3.337,
                "cost_per_test": 2.246783,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 85.08,
                "latency": 733.352,
                "stderr": 2.727,
                "cost_per_test": 0.291222,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 85.04,
                "latency": 6953.006,
                "stderr": 4.021,
                "cost_per_test": 4.449226,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 85.038,
                "latency": 6438.62,
                "stderr": 3.871,
                "cost_per_test": 22.403219,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 84.883,
                "latency": 355.395,
                "stderr": 2.734,
                "cost_per_test": 0.362114,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 84.836,
                "latency": 572.241,
                "stderr": 2.937,
                "cost_per_test": 2.490256,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 84.731,
                "latency": 1025.06,
                "stderr": 2.991,
                "cost_per_test": 0.204391,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 84.701,
                "latency": 1943.878,
                "stderr": 3.332,
                "cost_per_test": 0.828943,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 84.694,
                "latency": 275.608,
                "stderr": 4.064,
                "cost_per_test": 0.430216,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 84.322,
                "latency": 1681.88,
                "stderr": 3.584,
                "cost_per_test": 0.17967,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 84.319,
                "latency": 2858.564,
                "stderr": 3.335,
                "cost_per_test": 0.671944,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 84.16,
                "latency": 261.358,
                "stderr": 4.175,
                "cost_per_test": 0.485181,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 84.082,
                "latency": 214.833,
                "stderr": 2.738,
                "cost_per_test": 0.667444,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 84.029,
                "latency": 427.276,
                "stderr": 3.803,
                "cost_per_test": 2.058545,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 83.985,
                "latency": 1775.958,
                "stderr": 2.699,
                "cost_per_test": 2.692913,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 83.745,
                "latency": 2156.642,
                "stderr": 2.425,
                "cost_per_test": 5.731692,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 83.29,
                "latency": 522.986,
                "stderr": 4.048,
                "cost_per_test": 0.545149,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 83.156,
                "latency": 160.664,
                "stderr": 2.78,
                "cost_per_test": 0.393807,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 82.968,
                "latency": 3655.679,
                "stderr": 2.866,
                "cost_per_test": 6.783177,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 82.74,
                "latency": 611.367,
                "stderr": 3.508,
                "cost_per_test": 0.372278,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 82.728,
                "latency": 2834.861,
                "stderr": 3.541,
                "cost_per_test": 0.421978,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 82.71,
                "latency": 828.518,
                "stderr": 2.708,
                "cost_per_test": 0.208128,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 82.609,
                "latency": 1280.117,
                "stderr": 2.564,
                "cost_per_test": 8.37428,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 82.374,
                "latency": 303.387,
                "stderr": 3.028,
                "cost_per_test": 0.019445,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 82.294,
                "latency": 5507.901,
                "stderr": 3.843,
                "cost_per_test": 1.55544,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 82.248,
                "latency": 782.498,
                "stderr": 3.483,
                "cost_per_test": 1.136745,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 82.161,
                "latency": 384.09,
                "stderr": 3.376,
                "cost_per_test": 0.079592,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 82.013,
                "latency": 1286.448,
                "stderr": 3.51,
                "cost_per_test": 1.563366,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 81.8,
                "latency": 4486.623,
                "stderr": 4.575,
                "cost_per_test": 1.742342,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 81.271,
                "latency": 353.782,
                "stderr": 3.674,
                "cost_per_test": 0.052994,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 80.858,
                "latency": 589.79,
                "stderr": 3.171,
                "cost_per_test": 0.111862,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 80.153,
                "latency": 62.09,
                "stderr": 3.56,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 79.637,
                "latency": 223.214,
                "stderr": 2.872,
                "cost_per_test": 0.466165,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 79.224,
                "latency": 1564.212,
                "stderr": 3.12,
                "cost_per_test": 1.25835,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 78.962,
                "latency": 830.725,
                "stderr": 4.476,
                "cost_per_test": 0.076515,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 78.706,
                "latency": 696.299,
                "stderr": 3.142,
                "cost_per_test": 0.516518,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 78.044,
                "latency": 214.461,
                "stderr": 2.715,
                "cost_per_test": 0.119798,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 77.981,
                "latency": 3732.629,
                "stderr": 3.581,
                "cost_per_test": 2.673236,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 77.97,
                "latency": 1202.973,
                "stderr": 3.132,
                "cost_per_test": 4.442219,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 77.484,
                "latency": 61.408,
                "stderr": 3.778,
                "cost_per_test": 0.080105,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 77.373,
                "latency": 1958.935,
                "stderr": 4.237,
                "cost_per_test": 0.364867,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 76.763,
                "latency": 74.98,
                "stderr": 3.949,
                "cost_per_test": 0.047883,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 75.784,
                "latency": 352.653,
                "stderr": 3.82,
                "cost_per_test": 0.029482,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 74.824,
                "latency": 125.212,
                "stderr": 3.27,
                "cost_per_test": 0.250093,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 74.537,
                "latency": 6612.654,
                "stderr": 5.33,
                "cost_per_test": 7.25464,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 74.474,
                "latency": 276.602,
                "stderr": 4.007,
                "cost_per_test": 0.260695,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 74.368,
                "latency": 1326.72,
                "stderr": 3.648,
                "cost_per_test": 0.216972,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 74.113,
                "latency": 1433.795,
                "stderr": 3.484,
                "cost_per_test": 1.467472,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 73.093,
                "latency": 279.901,
                "stderr": 3.814,
                "cost_per_test": 0.255064,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 72.275,
                "latency": 176.967,
                "stderr": 4.471,
                "cost_per_test": 0.030415,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 70.15,
                "latency": 540.649,
                "stderr": 3.678,
                "cost_per_test": 0.230889,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 69.567,
                "latency": 449.168,
                "stderr": 4.186,
                "cost_per_test": 0.209051,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 69.251,
                "latency": 822.903,
                "stderr": 4.888,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 68.181,
                "latency": 715.501,
                "stderr": 4.487,
                "cost_per_test": 1.425655,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 65.258,
                "latency": 597.509,
                "stderr": 4.328,
                "cost_per_test": 0.134636,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 63.691,
                "latency": 293.416,
                "stderr": 4.815,
                "cost_per_test": 0.090182,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 62.892,
                "latency": 53.037,
                "stderr": 4.365,
                "cost_per_test": 0.058844,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 61.883,
                "latency": 21.603,
                "stderr": 4.315,
                "cost_per_test": 0.013343,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 53.161,
                "latency": 268.813,
                "stderr": 4.863,
                "cost_per_test": 0.003527,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__criminal": {
            "google/gemini-4-carbon": {
                "accuracy": 90.59,
                "latency": 3164.955,
                "stderr": 2.11,
                "cost_per_test": 8.842662,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 90.142,
                "latency": 1907.88,
                "stderr": 2.078,
                "cost_per_test": 6.616583,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 89.439,
                "latency": 1527.247,
                "stderr": 2.129,
                "cost_per_test": 3.617802,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 88.777,
                "latency": 315.871,
                "stderr": 2.213,
                "cost_per_test": 1.534611,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 88.075,
                "latency": 3466.552,
                "stderr": 2.124,
                "cost_per_test": 22.326482,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 88.073,
                "latency": 2118.512,
                "stderr": 2.29,
                "cost_per_test": 6.903486,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 87.949,
                "latency": 1371.988,
                "stderr": 2.265,
                "cost_per_test": 5.268708,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 87.747,
                "latency": 755.757,
                "stderr": 2.443,
                "cost_per_test": 2.852274,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 87.417,
                "latency": 967.654,
                "stderr": 2.225,
                "cost_per_test": 4.059119,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 87.238,
                "latency": 2957.873,
                "stderr": 2.506,
                "cost_per_test": 2.182135,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 87.179,
                "latency": 1356.005,
                "stderr": 2.392,
                "cost_per_test": 1.460147,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 87.174,
                "latency": 1928.013,
                "stderr": 2.313,
                "cost_per_test": 0.177786,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 86.657,
                "latency": 303.011,
                "stderr": 2.762,
                "cost_per_test": 0.60538,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 86.381,
                "latency": 4987.986,
                "stderr": 2.307,
                "cost_per_test": 24.12378,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 86.134,
                "latency": 1398.537,
                "stderr": 2.332,
                "cost_per_test": 9.468125,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 85.936,
                "latency": 2749.369,
                "stderr": 2.265,
                "cost_per_test": 0.250408,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 85.796,
                "latency": 989.26,
                "stderr": 2.204,
                "cost_per_test": 1.99149,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 85.453,
                "latency": 2183.678,
                "stderr": 2.39,
                "cost_per_test": 2.568801,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 85.445,
                "latency": 320.962,
                "stderr": 2.33,
                "cost_per_test": 0.345173,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 84.941,
                "latency": 3997.963,
                "stderr": 2.401,
                "cost_per_test": 2.676114,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 84.643,
                "latency": 772.67,
                "stderr": 2.462,
                "cost_per_test": 3.543085,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 84.532,
                "latency": 767.942,
                "stderr": 2.949,
                "cost_per_test": 0.321616,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 84.458,
                "latency": 2341.734,
                "stderr": 2.235,
                "cost_per_test": 3.38256,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 84.284,
                "latency": 347.571,
                "stderr": 2.652,
                "cost_per_test": 0.827264,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 84.249,
                "latency": 6259.782,
                "stderr": 2.412,
                "cost_per_test": 14.876209,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 84.145,
                "latency": 2210.107,
                "stderr": 2.586,
                "cost_per_test": 0.87429,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 84.113,
                "latency": 540.381,
                "stderr": 2.952,
                "cost_per_test": 0.616709,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 84.06,
                "latency": 1506.644,
                "stderr": 2.665,
                "cost_per_test": 2.518676,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 83.857,
                "latency": 2208.959,
                "stderr": 2.777,
                "cost_per_test": 0.046529,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 83.849,
                "latency": 5566.578,
                "stderr": 2.561,
                "cost_per_test": 1.725041,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 83.596,
                "latency": 2498.94,
                "stderr": 2.455,
                "cost_per_test": 1.017486,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 83.51,
                "latency": 5432.346,
                "stderr": 2.872,
                "cost_per_test": 3.124013,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 83.209,
                "latency": 2224.103,
                "stderr": 2.854,
                "cost_per_test": 0.715843,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 83.205,
                "latency": 1658.26,
                "stderr": 2.276,
                "cost_per_test": 10.941763,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 83.127,
                "latency": 301.211,
                "stderr": 3.178,
                "cost_per_test": 0.446419,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 82.909,
                "latency": 7162.351,
                "stderr": 3.481,
                "cost_per_test": 25.272629,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 82.619,
                "latency": 874.891,
                "stderr": 2.863,
                "cost_per_test": 0.822469,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 82.584,
                "latency": 5104.912,
                "stderr": 3.195,
                "cost_per_test": 2.104097,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 82.57,
                "latency": 2079.561,
                "stderr": 2.465,
                "cost_per_test": 3.601082,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 82.561,
                "latency": 4543.691,
                "stderr": 2.55,
                "cost_per_test": 8.912211,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 82.44,
                "latency": 256.075,
                "stderr": 2.626,
                "cost_per_test": 1.359179,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 82.222,
                "latency": 2563.313,
                "stderr": 2.562,
                "cost_per_test": 0.994665,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 82.029,
                "latency": 529.681,
                "stderr": 2.544,
                "cost_per_test": 2.779258,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 81.944,
                "latency": 7534.426,
                "stderr": 3.778,
                "cost_per_test": 5.274524,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 81.837,
                "latency": 2310.474,
                "stderr": 2.851,
                "cost_per_test": 7.322657,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 81.761,
                "latency": 3744.173,
                "stderr": 2.971,
                "cost_per_test": 0.819373,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 81.641,
                "latency": 425.103,
                "stderr": 2.748,
                "cost_per_test": 0.06767,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 81.349,
                "latency": 1712.774,
                "stderr": 2.933,
                "cost_per_test": 1.169956,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 81.01,
                "latency": 189.608,
                "stderr": 2.666,
                "cost_per_test": 0.501214,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 80.073,
                "latency": 1034.653,
                "stderr": 2.733,
                "cost_per_test": 0.846458,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 79.928,
                "latency": 2805.101,
                "stderr": 2.964,
                "cost_per_test": 0.425544,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 79.693,
                "latency": 1562.595,
                "stderr": 2.698,
                "cost_per_test": 5.495004,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 79.608,
                "latency": 1433.199,
                "stderr": 2.988,
                "cost_per_test": 1.995877,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 79.256,
                "latency": 682.99,
                "stderr": 3.087,
                "cost_per_test": 0.233636,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 79.004,
                "latency": 2463.873,
                "stderr": 3.058,
                "cost_per_test": 0.454694,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 78.143,
                "latency": 331.685,
                "stderr": 3.194,
                "cost_per_test": 0.405872,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 78.11,
                "latency": 855.631,
                "stderr": 2.891,
                "cost_per_test": 0.325282,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 78.109,
                "latency": 87.152,
                "stderr": 2.943,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 78.032,
                "latency": 711.836,
                "stderr": 2.961,
                "cost_per_test": 0.143249,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 77.693,
                "latency": 787.984,
                "stderr": 2.824,
                "cost_per_test": 1.332257,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 77.42,
                "latency": 872.79,
                "stderr": 2.901,
                "cost_per_test": 0.509253,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 76.836,
                "latency": 398.106,
                "stderr": 2.757,
                "cost_per_test": 0.081998,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 76.823,
                "latency": 1264.855,
                "stderr": 3.003,
                "cost_per_test": 0.25224,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 76.39,
                "latency": 1004.085,
                "stderr": 2.887,
                "cost_per_test": 0.615927,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 76.229,
                "latency": 4308.911,
                "stderr": 2.992,
                "cost_per_test": 3.298261,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 76.177,
                "latency": 217.466,
                "stderr": 2.892,
                "cost_per_test": 0.48486,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 75.057,
                "latency": 1937.218,
                "stderr": 2.901,
                "cost_per_test": 1.608029,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 74.773,
                "latency": 435.186,
                "stderr": 3.265,
                "cost_per_test": 0.022099,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 74.684,
                "latency": 1025.226,
                "stderr": 3.856,
                "cost_per_test": 0.071477,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 73.1,
                "latency": 858.129,
                "stderr": 3.258,
                "cost_per_test": 0.2621,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 72.798,
                "latency": 1398.791,
                "stderr": 3.182,
                "cost_per_test": 0.25486,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 72.137,
                "latency": 221.166,
                "stderr": 2.88,
                "cost_per_test": 0.13559,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 71.257,
                "latency": 78.654,
                "stderr": 3.4,
                "cost_per_test": 0.062754,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 69.682,
                "latency": 6680.114,
                "stderr": 4.745,
                "cost_per_test": 8.328752,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 69.316,
                "latency": 110.22,
                "stderr": 3.232,
                "cost_per_test": 0.199087,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 68.802,
                "latency": 698.746,
                "stderr": 3.45,
                "cost_per_test": 0.68798,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 68.108,
                "latency": 63.499,
                "stderr": 3.364,
                "cost_per_test": 0.122864,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 67.541,
                "latency": 394.953,
                "stderr": 3.311,
                "cost_per_test": 0.023143,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 67.321,
                "latency": 267.815,
                "stderr": 3.459,
                "cost_per_test": 0.155562,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 66.155,
                "latency": 651.134,
                "stderr": 3.844,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 64.416,
                "latency": 283.18,
                "stderr": 3.525,
                "cost_per_test": 0.260687,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 64.389,
                "latency": 2330.292,
                "stderr": 3.396,
                "cost_per_test": 2.024329,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 63.41,
                "latency": 421.649,
                "stderr": 3.619,
                "cost_per_test": 0.226204,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 62.879,
                "latency": 185.804,
                "stderr": 3.646,
                "cost_per_test": 0.034171,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 61.813,
                "latency": 48.522,
                "stderr": 3.438,
                "cost_per_test": 0.059734,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 61.647,
                "latency": 521.807,
                "stderr": 3.495,
                "cost_per_test": 0.26163,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 60.899,
                "latency": 611.807,
                "stderr": 3.853,
                "cost_per_test": 1.490522,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 57.92,
                "latency": 548.936,
                "stderr": 3.278,
                "cost_per_test": 0.153371,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 56.943,
                "latency": 20.102,
                "stderr": 3.825,
                "cost_per_test": 0.013186,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 56.382,
                "latency": 301.868,
                "stderr": 3.95,
                "cost_per_test": 0.108798,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 51.248,
                "latency": 187.227,
                "stderr": 3.656,
                "cost_per_test": 0.003839,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__family": {
            "anthropic/claude-opus-5": {
                "accuracy": 86.646,
                "latency": 1858.312,
                "stderr": 3.881,
                "cost_per_test": 6.195885,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 84.761,
                "latency": 4551.243,
                "stderr": 4.625,
                "cost_per_test": 12.648487,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 84.308,
                "latency": 357.472,
                "stderr": 4.498,
                "cost_per_test": 0.633898,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 83.42,
                "latency": 1437.652,
                "stderr": 4.044,
                "cost_per_test": 5.323844,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 83.048,
                "latency": 2250.577,
                "stderr": 4.509,
                "cost_per_test": 4.730031,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 81.509,
                "latency": 2821.52,
                "stderr": 4.4,
                "cost_per_test": 10.91868,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 80.388,
                "latency": 1405.167,
                "stderr": 3.936,
                "cost_per_test": 1.425494,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 80.107,
                "latency": 2827.87,
                "stderr": 5.612,
                "cost_per_test": 0.051257,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 79.482,
                "latency": 855.894,
                "stderr": 3.788,
                "cost_per_test": 0.383913,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 79.323,
                "latency": 5063.727,
                "stderr": 3.94,
                "cost_per_test": 9.812737,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 79.009,
                "latency": 716.66,
                "stderr": 4.403,
                "cost_per_test": 0.822081,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 78.4,
                "latency": 3017.714,
                "stderr": 4.161,
                "cost_per_test": 1.168665,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 78.125,
                "latency": 3202.004,
                "stderr": 5.529,
                "cost_per_test": 0.286053,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 77.897,
                "latency": 1643.011,
                "stderr": 5.851,
                "cost_per_test": 11.011763,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 77.732,
                "latency": 1013.398,
                "stderr": 4.697,
                "cost_per_test": 0.323772,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 77.576,
                "latency": 3849.877,
                "stderr": 5.756,
                "cost_per_test": 2.697912,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 77.545,
                "latency": 1482.157,
                "stderr": 4.578,
                "cost_per_test": 5.56694,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 77.423,
                "latency": 2649.862,
                "stderr": 5.609,
                "cost_per_test": 0.231872,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 77.415,
                "latency": 3235.665,
                "stderr": 6.068,
                "cost_per_test": 21.529237,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 77.323,
                "latency": 787.574,
                "stderr": 5.197,
                "cost_per_test": 2.857814,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 77.307,
                "latency": 1896.183,
                "stderr": 3.995,
                "cost_per_test": 15.003346,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 77.267,
                "latency": 3456.123,
                "stderr": 4.381,
                "cost_per_test": 1.536833,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 77.072,
                "latency": 5320.428,
                "stderr": 4.982,
                "cost_per_test": 12.577804,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 76.751,
                "latency": 1189.264,
                "stderr": 4.6,
                "cost_per_test": 2.408033,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 76.736,
                "latency": 6569.663,
                "stderr": 4.515,
                "cost_per_test": 30.084962,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 76.37,
                "latency": 8086.246,
                "stderr": 6.633,
                "cost_per_test": 5.806978,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 76.357,
                "latency": 6312.514,
                "stderr": 4.952,
                "cost_per_test": 3.697863,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 76.317,
                "latency": 638.032,
                "stderr": 4.805,
                "cost_per_test": 3.584147,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 75.432,
                "latency": 4757.56,
                "stderr": 5.053,
                "cost_per_test": 3.063697,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 75.4,
                "latency": 3035.777,
                "stderr": 5.835,
                "cost_per_test": 5.439212,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 75.195,
                "latency": 4670.091,
                "stderr": 5.246,
                "cost_per_test": 0.970682,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 75.03,
                "latency": 613.145,
                "stderr": 6.035,
                "cost_per_test": 2.522005,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 74.889,
                "latency": 306.75,
                "stderr": 4.464,
                "cost_per_test": 1.594003,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 74.466,
                "latency": 328.141,
                "stderr": 5.024,
                "cost_per_test": 0.452803,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 74.285,
                "latency": 1762.575,
                "stderr": 4.898,
                "cost_per_test": 0.33418,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 74.191,
                "latency": 2445.946,
                "stderr": 5.801,
                "cost_per_test": 0.826784,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 74.056,
                "latency": 5152.106,
                "stderr": 5.192,
                "cost_per_test": 1.724456,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 73.686,
                "latency": 348.059,
                "stderr": 4.833,
                "cost_per_test": 0.406411,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 73.433,
                "latency": 2357.557,
                "stderr": 4.48,
                "cost_per_test": 3.772956,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 73.24,
                "latency": 6801.677,
                "stderr": 5.577,
                "cost_per_test": 2.756433,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 72.796,
                "latency": 2019.893,
                "stderr": 5.301,
                "cost_per_test": 0.418234,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 72.424,
                "latency": 966.329,
                "stderr": 5.56,
                "cost_per_test": 4.336354,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 72.268,
                "latency": 2332.251,
                "stderr": 4.968,
                "cost_per_test": 1.856297,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 72.233,
                "latency": 1564.903,
                "stderr": 5.607,
                "cost_per_test": 2.758903,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 71.694,
                "latency": 1101.097,
                "stderr": 5.673,
                "cost_per_test": 0.562977,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 71.523,
                "latency": 2917.375,
                "stderr": 4.983,
                "cost_per_test": 0.465031,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 71.357,
                "latency": 3983.979,
                "stderr": 5.35,
                "cost_per_test": 0.672496,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 70.901,
                "latency": 1329.086,
                "stderr": 5.57,
                "cost_per_test": 0.070993,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 70.836,
                "latency": 2680.585,
                "stderr": 4.872,
                "cost_per_test": 7.622096,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 70.56,
                "latency": 839.919,
                "stderr": 5.33,
                "cost_per_test": 0.991333,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 70.504,
                "latency": 2391.966,
                "stderr": 5.948,
                "cost_per_test": 0.88974,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 70.349,
                "latency": 121.037,
                "stderr": 5.333,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 69.315,
                "latency": 1806.747,
                "stderr": 6.228,
                "cost_per_test": 1.50639,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 69.188,
                "latency": 390.844,
                "stderr": 6.004,
                "cost_per_test": 0.986437,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 69.17,
                "latency": 1713.026,
                "stderr": 6.215,
                "cost_per_test": 2.549692,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 69.065,
                "latency": 1758.115,
                "stderr": 5.533,
                "cost_per_test": 5.389618,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 68.892,
                "latency": 1001.225,
                "stderr": 4.477,
                "cost_per_test": 0.173654,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 68.039,
                "latency": 951.192,
                "stderr": 5.235,
                "cost_per_test": 0.370092,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 67.53,
                "latency": 514.571,
                "stderr": 6.551,
                "cost_per_test": 0.08758,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 66.252,
                "latency": 1317.912,
                "stderr": 5.618,
                "cost_per_test": 0.816394,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 66.028,
                "latency": 7441.78,
                "stderr": 7.454,
                "cost_per_test": 25.835741,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 64.806,
                "latency": 4690.594,
                "stderr": 5.881,
                "cost_per_test": 3.506565,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 63.564,
                "latency": 6905.045,
                "stderr": 8.026,
                "cost_per_test": 8.216145,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 62.107,
                "latency": 167.284,
                "stderr": 6.045,
                "cost_per_test": 0.168183,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 61.607,
                "latency": 1722.37,
                "stderr": 6.713,
                "cost_per_test": 0.30233,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 60.959,
                "latency": 2863.795,
                "stderr": 6.121,
                "cost_per_test": 2.335677,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 59.646,
                "latency": 727.744,
                "stderr": 6.04,
                "cost_per_test": 1.001303,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 59.585,
                "latency": 195.123,
                "stderr": 6.152,
                "cost_per_test": 0.581541,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 59.268,
                "latency": 900.57,
                "stderr": 6.671,
                "cost_per_test": 1.420719,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 58.608,
                "latency": 1055.917,
                "stderr": 6.891,
                "cost_per_test": 0.443026,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 58.209,
                "latency": 434.377,
                "stderr": 6.265,
                "cost_per_test": 4.138935,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 58.154,
                "latency": 342.126,
                "stderr": 5.699,
                "cost_per_test": 0.41232,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 58.118,
                "latency": 435.98,
                "stderr": 5.723,
                "cost_per_test": 0.075731,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 53.768,
                "latency": 98.358,
                "stderr": 6.267,
                "cost_per_test": 0.074199,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 52.649,
                "latency": 422.665,
                "stderr": 6.426,
                "cost_per_test": 0.030982,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 51.571,
                "latency": 692.635,
                "stderr": 6.939,
                "cost_per_test": 0.039086,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 51.57,
                "latency": 4205.69,
                "stderr": 5.858,
                "cost_per_test": 2.933958,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 51.562,
                "latency": 458.965,
                "stderr": 6.465,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 50.406,
                "latency": 250.897,
                "stderr": 6.917,
                "cost_per_test": 0.288946,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 49.711,
                "latency": 62.256,
                "stderr": 6.034,
                "cost_per_test": 0.169797,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 47.346,
                "latency": 86.589,
                "stderr": 6.258,
                "cost_per_test": 0.230167,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 47.229,
                "latency": 222.566,
                "stderr": 6.364,
                "cost_per_test": 0.036939,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 45.767,
                "latency": 436.535,
                "stderr": 6.323,
                "cost_per_test": 0.255715,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 43.293,
                "latency": 592.178,
                "stderr": 6.228,
                "cost_per_test": 0.140683,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 43.161,
                "latency": 246.131,
                "stderr": 5.728,
                "cost_per_test": 0.16395,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 42.725,
                "latency": 570.598,
                "stderr": 5.244,
                "cost_per_test": 1.090579,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 41.993,
                "latency": 532.251,
                "stderr": 5.847,
                "cost_per_test": 0.261764,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 40.393,
                "latency": 311.615,
                "stderr": 5.049,
                "cost_per_test": 0.005312,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 37.561,
                "latency": 47.19,
                "stderr": 5.479,
                "cost_per_test": 0.062993,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 36.546,
                "latency": 18.535,
                "stderr": 5.492,
                "cost_per_test": 0.013412,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 30.22,
                "latency": 383.551,
                "stderr": 5.051,
                "cost_per_test": 0.13236,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            }
        },
        "weighted_pass__health": {
            "openai/gpt-5.6-sol": {
                "accuracy": 97.511,
                "latency": 4151.253,
                "stderr": 1.194,
                "cost_per_test": 22.836198,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 96.79,
                "latency": 1346.948,
                "stderr": 1.387,
                "cost_per_test": 7.009258,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 96.764,
                "latency": 2450.538,
                "stderr": 1.293,
                "cost_per_test": 6.751201,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 96.691,
                "latency": 1693.184,
                "stderr": 1.324,
                "cost_per_test": 1.828426,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 96.459,
                "latency": 858.216,
                "stderr": 1.845,
                "cost_per_test": 4.579894,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 96.433,
                "latency": 4630.853,
                "stderr": 1.922,
                "cost_per_test": 12.659413,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 96.193,
                "latency": 2538.105,
                "stderr": 1.602,
                "cost_per_test": 4.232271,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 96.178,
                "latency": 1923.984,
                "stderr": 1.399,
                "cost_per_test": 0.708547,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 96.166,
                "latency": 3496.535,
                "stderr": 1.354,
                "cost_per_test": 2.5406,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 96.042,
                "latency": 1338.036,
                "stderr": 1.567,
                "cost_per_test": 4.625213,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 95.75,
                "latency": 1838.285,
                "stderr": 1.935,
                "cost_per_test": 3.207198,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 95.589,
                "latency": 710.57,
                "stderr": 1.643,
                "cost_per_test": 0.327988,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 95.533,
                "latency": 1095.281,
                "stderr": 1.692,
                "cost_per_test": 4.723349,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 95.5,
                "latency": 1414.958,
                "stderr": 1.758,
                "cost_per_test": 3.858235,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 95.48,
                "latency": 3604.391,
                "stderr": 1.779,
                "cost_per_test": 0.88574,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 95.169,
                "latency": 3215.749,
                "stderr": 2.39,
                "cost_per_test": 2.470231,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-opus-5-5": {
                "accuracy": 95.001,
                "latency": 5675.961,
                "stderr": 3.385,
                "cost_per_test": 20.596615,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 94.942,
                "latency": 1360.302,
                "stderr": 1.89,
                "cost_per_test": 0.193071,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 94.925,
                "latency": 706.355,
                "stderr": 2.036,
                "cost_per_test": 0.272236,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 94.899,
                "latency": 1553.826,
                "stderr": 2.096,
                "cost_per_test": 0.926438,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 94.798,
                "latency": 986.856,
                "stderr": 2.149,
                "cost_per_test": 10.254175,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 94.599,
                "latency": 836.886,
                "stderr": 2.138,
                "cost_per_test": 4.339603,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 94.553,
                "latency": 318.119,
                "stderr": 2.621,
                "cost_per_test": 0.64171,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 94.52,
                "latency": 470.162,
                "stderr": 2.646,
                "cost_per_test": 0.708927,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 94.516,
                "latency": 1126.576,
                "stderr": 1.95,
                "cost_per_test": 12.38182,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 94.49,
                "latency": 3508.124,
                "stderr": 1.842,
                "cost_per_test": 1.530774,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 94.337,
                "latency": 1922.826,
                "stderr": 2.045,
                "cost_per_test": 2.642852,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 94.319,
                "latency": 2681.332,
                "stderr": 1.839,
                "cost_per_test": 0.057691,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 93.826,
                "latency": 2414.0,
                "stderr": 2.285,
                "cost_per_test": 0.809076,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 93.821,
                "latency": 282.896,
                "stderr": 2.095,
                "cost_per_test": 1.899416,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 93.698,
                "latency": 3030.813,
                "stderr": 3.593,
                "cost_per_test": 22.907678,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 93.59,
                "latency": 2062.802,
                "stderr": 2.623,
                "cost_per_test": 9.360291,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 93.249,
                "latency": 1266.05,
                "stderr": 2.415,
                "cost_per_test": 2.204201,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 93.206,
                "latency": 3850.231,
                "stderr": 3.138,
                "cost_per_test": 8.132992,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 92.926,
                "latency": 3278.164,
                "stderr": 2.635,
                "cost_per_test": 0.238934,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 92.925,
                "latency": 303.781,
                "stderr": 2.219,
                "cost_per_test": 0.574895,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 92.819,
                "latency": 3359.798,
                "stderr": 2.602,
                "cost_per_test": 3.11111,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 92.435,
                "latency": 5897.488,
                "stderr": 2.313,
                "cost_per_test": 2.500509,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 92.161,
                "latency": 800.16,
                "stderr": 2.489,
                "cost_per_test": 0.394232,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 92.146,
                "latency": 1364.3,
                "stderr": 2.478,
                "cost_per_test": 2.657382,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 91.692,
                "latency": 1026.417,
                "stderr": 2.536,
                "cost_per_test": 3.793295,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 91.474,
                "latency": 274.578,
                "stderr": 1.976,
                "cost_per_test": 1.402227,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 91.257,
                "latency": 931.988,
                "stderr": 2.374,
                "cost_per_test": 0.931196,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 91.229,
                "latency": 999.255,
                "stderr": 2.203,
                "cost_per_test": 0.81017,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 91.114,
                "latency": 2156.169,
                "stderr": 3.007,
                "cost_per_test": 0.374147,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 90.816,
                "latency": 663.425,
                "stderr": 2.648,
                "cost_per_test": 3.159294,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 90.803,
                "latency": 1685.594,
                "stderr": 2.474,
                "cost_per_test": 1.561613,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 90.529,
                "latency": 245.802,
                "stderr": 2.623,
                "cost_per_test": 0.701623,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 90.496,
                "latency": 5711.873,
                "stderr": 4.683,
                "cost_per_test": 6.862578,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 90.283,
                "latency": 1047.406,
                "stderr": 2.565,
                "cost_per_test": 3.45972,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 90.249,
                "latency": 97.234,
                "stderr": 2.59,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 89.408,
                "latency": 2028.962,
                "stderr": 3.661,
                "cost_per_test": 0.905473,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 89.275,
                "latency": 2705.795,
                "stderr": 3.647,
                "cost_per_test": 1.314815,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 88.772,
                "latency": 7418.628,
                "stderr": 4.698,
                "cost_per_test": 5.193942,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 88.551,
                "latency": 472.047,
                "stderr": 2.913,
                "cost_per_test": 0.076792,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 88.485,
                "latency": 910.936,
                "stderr": 2.857,
                "cost_per_test": 0.496174,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 88.212,
                "latency": 462.415,
                "stderr": 3.469,
                "cost_per_test": 4.040025,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 87.936,
                "latency": 298.421,
                "stderr": 2.581,
                "cost_per_test": 0.506452,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 87.857,
                "latency": 134.685,
                "stderr": 2.664,
                "cost_per_test": 0.416767,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 87.356,
                "latency": 781.127,
                "stderr": 4.19,
                "cost_per_test": 0.081431,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 87.176,
                "latency": 3297.206,
                "stderr": 3.353,
                "cost_per_test": 0.553927,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 86.824,
                "latency": 145.057,
                "stderr": 3.081,
                "cost_per_test": 0.136684,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 86.658,
                "latency": 848.765,
                "stderr": 3.085,
                "cost_per_test": 1.786099,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 86.614,
                "latency": 1384.623,
                "stderr": 2.951,
                "cost_per_test": 0.400672,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 85.452,
                "latency": 830.651,
                "stderr": 3.503,
                "cost_per_test": 0.505144,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 84.385,
                "latency": 998.171,
                "stderr": 3.091,
                "cost_per_test": 0.155212,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 84.258,
                "latency": 5571.399,
                "stderr": 6.239,
                "cost_per_test": 2.397715,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 83.436,
                "latency": 232.285,
                "stderr": 3.06,
                "cost_per_test": 1.804209,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 82.755,
                "latency": 322.342,
                "stderr": 3.924,
                "cost_per_test": 0.321248,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 81.669,
                "latency": 1608.673,
                "stderr": 3.886,
                "cost_per_test": 1.371176,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 81.212,
                "latency": 79.862,
                "stderr": 3.744,
                "cost_per_test": 0.233416,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 80.779,
                "latency": 387.77,
                "stderr": 4.363,
                "cost_per_test": 0.119406,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 80.726,
                "latency": 477.359,
                "stderr": 3.916,
                "cost_per_test": 0.028475,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 79.444,
                "latency": 696.803,
                "stderr": 4.502,
                "cost_per_test": 0.848896,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 75.743,
                "latency": 48.195,
                "stderr": 3.902,
                "cost_per_test": 0.13408,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 74.397,
                "latency": 1152.919,
                "stderr": 4.038,
                "cost_per_test": 0.18555,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 73.839,
                "latency": 71.497,
                "stderr": 3.946,
                "cost_per_test": 0.055137,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 73.772,
                "latency": 1671.994,
                "stderr": 4.948,
                "cost_per_test": 1.329631,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 73.53,
                "latency": 538.653,
                "stderr": 3.833,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 72.188,
                "latency": 334.244,
                "stderr": 4.417,
                "cost_per_test": 0.106495,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 71.555,
                "latency": 372.619,
                "stderr": 4.086,
                "cost_per_test": 0.056492,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 69.817,
                "latency": 232.734,
                "stderr": 4.727,
                "cost_per_test": 0.036118,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 69.597,
                "latency": 587.499,
                "stderr": 4.538,
                "cost_per_test": 0.21973,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 69.187,
                "latency": 474.816,
                "stderr": 3.476,
                "cost_per_test": 0.199185,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 67.923,
                "latency": 188.923,
                "stderr": 4.367,
                "cost_per_test": 0.290063,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 67.789,
                "latency": 618.768,
                "stderr": 4.381,
                "cost_per_test": 1.063116,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 62.795,
                "latency": 396.297,
                "stderr": 4.13,
                "cost_per_test": 0.09495,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 62.305,
                "latency": 229.682,
                "stderr": 4.367,
                "cost_per_test": 0.532775,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 57.555,
                "latency": 51.629,
                "stderr": 4.978,
                "cost_per_test": 0.048442,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 56.579,
                "latency": 19.508,
                "stderr": 4.937,
                "cost_per_test": 0.013992,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 45.376,
                "latency": 661.96,
                "stderr": 4.286,
                "cost_per_test": 0.005274,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            }
        },
        "weighted_pass__immigration": {
            "anthropic/claude-opus-5-5": {
                "accuracy": 90.506,
                "latency": 7783.489,
                "stderr": 3.843,
                "cost_per_test": 28.846858,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-argon": {
                "accuracy": 89.867,
                "latency": 1545.533,
                "stderr": 4.933,
                "cost_per_test": 4.782473,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-fable-5-1": {
                "accuracy": 88.941,
                "latency": 4229.709,
                "stderr": 5.791,
                "cost_per_test": 29.748136,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-4-carbon": {
                "accuracy": 87.445,
                "latency": 2577.035,
                "stderr": 4.701,
                "cost_per_test": 7.345528,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 1048576,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "anthropic/claude-opus-5": {
                "accuracy": 86.928,
                "latency": 2091.212,
                "stderr": 6.787,
                "cost_per_test": 8.705325,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "grok/grok-4.6": {
                "accuracy": 85.354,
                "latency": 2227.72,
                "stderr": 7.505,
                "cost_per_test": 2.139853,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "meta/muse-wonder-full": {
                "accuracy": 85.35,
                "latency": 2309.031,
                "stderr": 6.259,
                "cost_per_test": 5.522954,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.7": {
                "accuracy": 85.296,
                "latency": 1309.204,
                "stderr": 6.529,
                "cost_per_test": 5.041009,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-6-astra": {
                "accuracy": 84.347,
                "latency": 2079.496,
                "stderr": 5.891,
                "cost_per_test": 14.716249,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "meta/muse_spark_1_2": {
                "accuracy": 84.075,
                "latency": 372.58,
                "stderr": 7.185,
                "cost_per_test": 1.011872,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.7-flash": {
                "accuracy": 83.773,
                "latency": 145.322,
                "stderr": 6.437,
                "cost_per_test": 0.576237,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "zai/glm-legal": {
                "accuracy": 83.657,
                "latency": 2188.906,
                "stderr": 7.29,
                "cost_per_test": 0.050144,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 83.475,
                "latency": 6121.239,
                "stderr": 8.164,
                "cost_per_test": 15.318633,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-large-4": {
                "accuracy": 83.421,
                "latency": 5257.651,
                "stderr": 5.767,
                "cost_per_test": 3.326554,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "anthropic/claude-opus-4-7": {
                "accuracy": 83.175,
                "latency": 522.726,
                "stderr": 7.265,
                "cost_per_test": 3.815712,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 83.117,
                "latency": 5391.649,
                "stderr": 7.066,
                "cost_per_test": 2.04155,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "google/gemini-3.8-flash": {
                "accuracy": 82.857,
                "latency": 231.193,
                "stderr": 6.336,
                "cost_per_test": 1.096453,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "meta/muse_spark_1_1": {
                "accuracy": 82.55,
                "latency": 305.943,
                "stderr": 6.309,
                "cost_per_test": 0.448482,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 82.311,
                "latency": 771.306,
                "stderr": 7.186,
                "cost_per_test": 1.548677,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "openai/gpt-5.5": {
                "accuracy": 82.114,
                "latency": 2512.615,
                "stderr": 6.489,
                "cost_per_test": 8.597148,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.3": {
                "accuracy": 82.025,
                "latency": 4767.048,
                "stderr": 6.62,
                "cost_per_test": 3.223316,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "thinkingmachines/inkling-2-small": {
                "accuracy": 81.821,
                "latency": 733.81,
                "stderr": 6.163,
                "cost_per_test": 0.307991,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "kimi/kimi-k3": {
                "accuracy": 81.495,
                "latency": 852.202,
                "stderr": 7.888,
                "cost_per_test": 4.505049,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "thinkingmachines/inkling": {
                "accuracy": 81.289,
                "latency": 1086.925,
                "stderr": 7.018,
                "cost_per_test": 0.562099,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 81.2,
                "latency": 1055.89,
                "stderr": 7.615,
                "cost_per_test": 4.615585,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "openai/gpt-5.6-sol": {
                "accuracy": 81.155,
                "latency": 5565.396,
                "stderr": 7.633,
                "cost_per_test": 29.782421,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-fable-5": {
                "accuracy": 80.986,
                "latency": 1290.179,
                "stderr": 7.784,
                "cost_per_test": 9.559148,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.1": {
                "accuracy": 80.706,
                "latency": 389.141,
                "stderr": 7.246,
                "cost_per_test": 0.560419,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "openai/gpt-5.6-terra": {
                "accuracy": 80.694,
                "latency": 5146.748,
                "stderr": 6.942,
                "cost_per_test": 9.402071,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "deepseek/deepseek-v4-pro-0813": {
                "accuracy": 80.657,
                "latency": 2999.169,
                "stderr": 7.595,
                "cost_per_test": 1.324131,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "fireworks/ember-1": {
                "accuracy": 80.576,
                "latency": 2203.778,
                "stderr": 6.556,
                "cost_per_test": 2.644131,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "meta/muse_spark_1_3_max": {
                "accuracy": 80.509,
                "latency": 377.453,
                "stderr": 6.956,
                "cost_per_test": 0.749015,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "meta/muse_spark_1_3": {
                "accuracy": 80.332,
                "latency": 586.852,
                "stderr": 7.907,
                "cost_per_test": 0.921916,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Meta",
                "harness": null
            },
            "google/gemini-3.5-flash": {
                "accuracy": 80.189,
                "latency": 219.971,
                "stderr": 6.859,
                "cost_per_test": 0.884226,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3.6-flash": {
                "accuracy": 80.097,
                "latency": 147.165,
                "stderr": 7.353,
                "cost_per_test": 0.405548,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 79.384,
                "latency": 3943.62,
                "stderr": 8.232,
                "cost_per_test": 3.006031,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 79.236,
                "latency": 3025.432,
                "stderr": 6.519,
                "cost_per_test": 5.242893,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "anthropic/claude-sonnet-5": {
                "accuracy": 79.199,
                "latency": 1629.748,
                "stderr": 7.413,
                "cost_per_test": 2.754829,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash-1017": {
                "accuracy": 79.031,
                "latency": 2363.885,
                "stderr": 7.593,
                "cost_per_test": 0.899309,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "alibaba/qwen3.6-plus": {
                "accuracy": 78.75,
                "latency": 598.99,
                "stderr": 8.816,
                "cost_per_test": 0.650754,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-6-luna": {
                "accuracy": 78.594,
                "latency": 3576.368,
                "stderr": 8.935,
                "cost_per_test": 0.63176,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "minimax/MiniMax-M3": {
                "accuracy": 78.593,
                "latency": 882.27,
                "stderr": 8.748,
                "cost_per_test": 0.482243,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 512000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-5.6-luna": {
                "accuracy": 78.572,
                "latency": 2966.232,
                "stderr": 7.434,
                "cost_per_test": 1.017586,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "zai/glm-5.3-flash": {
                "accuracy": 78.447,
                "latency": 4168.205,
                "stderr": 8.598,
                "cost_per_test": 0.281372,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "anthropic/claude-opus-4-8": {
                "accuracy": 78.35,
                "latency": 728.3,
                "stderr": 8.403,
                "cost_per_test": 3.441413,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "zai/glm-5.2": {
                "accuracy": 78.068,
                "latency": 1192.0,
                "stderr": 8.926,
                "cost_per_test": 1.003117,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Zhipu AI",
                "harness": null
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 78.046,
                "latency": 2445.483,
                "stderr": 7.935,
                "cost_per_test": 3.066554,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "anthropic/claude-sonnet-4-6": {
                "accuracy": 77.81,
                "latency": 1048.83,
                "stderr": 8.892,
                "cost_per_test": 2.703862,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": true,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "Anthropic",
                "harness": null
            },
            "xiaomi/mimo-v2.5-pro": {
                "accuracy": 77.663,
                "latency": 376.011,
                "stderr": 8.837,
                "cost_per_test": 0.081151,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "stepfun/step-5-preview": {
                "accuracy": 77.592,
                "latency": 2795.76,
                "stderr": 8.491,
                "cost_per_test": 1.125625,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 1024000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Stepfun",
                "harness": null
            },
            "deepseek/deepseek-v4.1-flash": {
                "accuracy": 77.311,
                "latency": 1059.33,
                "stderr": 9.055,
                "cost_per_test": 0.440884,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "kimi/kimi-k2.5-thinking": {
                "accuracy": 77.0,
                "latency": 866.165,
                "stderr": 7.278,
                "cost_per_test": 0.178499,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "minimax/MiniMax-M3.1-flash": {
                "accuracy": 76.913,
                "latency": 3289.995,
                "stderr": 7.373,
                "cost_per_test": 0.523583,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": "max",
                "provider": "MiniMax",
                "harness": null
            },
            "meta/muse-spark-open": {
                "accuracy": 76.745,
                "latency": 1537.666,
                "stderr": 8.465,
                "cost_per_test": 0.570169,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Together AI",
                "harness": null
            },
            "xiaomi/mimo-v2.6-pro": {
                "accuracy": 76.548,
                "latency": 1793.408,
                "stderr": 8.224,
                "cost_per_test": 0.234002,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "deepseek/deepseek-v4-flash-0731": {
                "accuracy": 76.442,
                "latency": 1260.914,
                "stderr": 8.243,
                "cost_per_test": 0.296663,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "tencent/hy4": {
                "accuracy": 76.387,
                "latency": 6986.35,
                "stderr": 9.526,
                "cost_per_test": 2.97227,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "openai/gpt-6-sol": {
                "accuracy": 76.217,
                "latency": 1914.142,
                "stderr": 7.996,
                "cost_per_test": 6.6645,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 76.087,
                "latency": 812.146,
                "stderr": 8.196,
                "cost_per_test": 0.150825,
                "temperature": 1.0,
                "top_p": 1.0,
                "max_output_tokens": 262000,
                "reasoning": null,
                "reasoning_effort": "0.99",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Thinkingmachines",
                "harness": null
            },
            "anthropic/claude-haiku-4-5-20251001-thinking": {
                "accuracy": 75.953,
                "latency": 176.78,
                "stderr": 7.7,
                "cost_per_test": 0.283317,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Anthropic",
                "harness": null
            },
            "mistralai/mistral-large-4-oss": {
                "accuracy": 75.908,
                "latency": 2067.692,
                "stderr": 9.428,
                "cost_per_test": 0.952254,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 256000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "inception/mercury-3": {
                "accuracy": 75.725,
                "latency": 449.603,
                "stderr": 7.536,
                "cost_per_test": 0.076601,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "deepseek/deepseek-v4-pro": {
                "accuracy": 75.593,
                "latency": 902.658,
                "stderr": 8.908,
                "cost_per_test": 0.700549,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 384000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "kimi/kimi-k2.6": {
                "accuracy": 75.554,
                "latency": 1561.245,
                "stderr": 8.508,
                "cost_per_test": 1.49714,
                "temperature": null,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "tencent/hy4-preview": {
                "accuracy": 75.048,
                "latency": 4515.312,
                "stderr": 8.147,
                "cost_per_test": 1.089,
                "temperature": 0.9,
                "top_p": 1.0,
                "max_output_tokens": 64000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Tencent",
                "harness": null
            },
            "ant/ling-3.0-flash-af-rc3": {
                "accuracy": 74.597,
                "latency": 659.266,
                "stderr": 8.528,
                "cost_per_test": 0.026153,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "xiaomi/mimo-v2.6-flash": {
                "accuracy": 74.549,
                "latency": 1272.83,
                "stderr": 7.195,
                "cost_per_test": 0.124704,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 74.064,
                "latency": 851.042,
                "stderr": 8.604,
                "cost_per_test": 1.91176,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Mistral AI",
                "harness": null
            },
            "reflection/Beam-1.0": {
                "accuracy": 73.84,
                "latency": 71.817,
                "stderr": 8.127,
                "cost_per_test": null,
                "temperature": 0.7,
                "top_p": 0.99,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Reflection",
                "harness": null
            },
            "nvidia/nemotron-3-ultra-550b-a55b": {
                "accuracy": 73.479,
                "latency": 848.642,
                "stderr": 9.162,
                "cost_per_test": null,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Nvidia",
                "harness": null
            },
            "alibaba/qwen3.7-max": {
                "accuracy": 73.271,
                "latency": 705.891,
                "stderr": 9.271,
                "cost_per_test": 1.495632,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "google/gemini-3.1-pro-preview": {
                "accuracy": 72.894,
                "latency": 185.396,
                "stderr": 9.888,
                "cost_per_test": 0.389491,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "google/gemini-3-flash-preview": {
                "accuracy": 71.531,
                "latency": 128.854,
                "stderr": 8.032,
                "cost_per_test": 0.088481,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "minimax/MiniMax-M2.7": {
                "accuracy": 69.874,
                "latency": 356.501,
                "stderr": 8.499,
                "cost_per_test": 0.229174,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 80000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "MiniMax",
                "harness": null
            },
            "openai/gpt-6.1-terra": {
                "accuracy": 69.793,
                "latency": 5272.469,
                "stderr": 9.655,
                "cost_per_test": 3.88229,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "google/gemini-3.5-flash-lite": {
                "accuracy": 68.693,
                "latency": 60.521,
                "stderr": 9.72,
                "cost_per_test": 0.102995,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "openai/gpt-5.4-mini-2026-03-17": {
                "accuracy": 68.382,
                "latency": 2027.42,
                "stderr": 8.841,
                "cost_per_test": 1.631269,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "xhigh",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "baseten/nemotron-3.5-super-vl-120b-a12b": {
                "accuracy": 68.213,
                "latency": 69.703,
                "stderr": 7.787,
                "cost_per_test": 0.06269,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Baseten",
                "harness": null
            },
            "grok/grok-4.3": {
                "accuracy": 67.726,
                "latency": 242.659,
                "stderr": 8.228,
                "cost_per_test": 0.366775,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "alibaba/qwen3.7-plus": {
                "accuracy": 66.867,
                "latency": 823.783,
                "stderr": 10.101,
                "cost_per_test": 0.603054,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "openai/gpt-5.4-nano-2026-03-17": {
                "accuracy": 65.25,
                "latency": 559.812,
                "stderr": 7.857,
                "cost_per_test": 0.159001,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "OpenAI",
                "harness": null
            },
            "inception/mercury-2.5": {
                "accuracy": 64.376,
                "latency": 38.568,
                "stderr": 7.908,
                "cost_per_test": 0.046126,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Inception",
                "harness": null
            },
            "poolside/laguna-m.1": {
                "accuracy": 64.305,
                "latency": 610.78,
                "stderr": 8.96,
                "cost_per_test": 0.296042,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "ant/ling-3.0-flash-2607": {
                "accuracy": 61.671,
                "latency": 201.446,
                "stderr": 9.534,
                "cost_per_test": 0.033288,
                "temperature": 0.6,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Ant",
                "harness": null
            },
            "grok/grok-4.20-0309-reasoning": {
                "accuracy": 60.533,
                "latency": 255.303,
                "stderr": 9.866,
                "cost_per_test": 0.300956,
                "temperature": 0.7,
                "top_p": 0.95,
                "max_output_tokens": 2000000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "SpaceXAI",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 57.38,
                "latency": 473.789,
                "stderr": 8.96,
                "cost_per_test": 0.036168,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Xiaomi",
                "harness": null
            },
            "google/gemini-3.1-flash-lite-preview": {
                "accuracy": 57.024,
                "latency": 21.669,
                "stderr": 9.462,
                "cost_per_test": 0.014207,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "alibaba/qwen3.8-max-internal": {
                "accuracy": 53.543,
                "latency": 7303.148,
                "stderr": 13.701,
                "cost_per_test": 4.997493,
                "temperature": 0.7,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Alibaba",
                "harness": null
            },
            "poolside/laguna-xs.2": {
                "accuracy": 51.559,
                "latency": 394.752,
                "stderr": 9.455,
                "cost_per_test": 0.099916,
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 32768,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Poolside",
                "harness": null
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 50.44,
                "latency": 173.024,
                "stderr": 9.552,
                "cost_per_test": 0.003083,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 131072,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Fireworks AI",
                "harness": null
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 32.709,
                "latency": 7116.076,
                "stderr": 12.681,
                "cost_per_test": 8.488087,
                "temperature": 1.0,
                "top_p": 0.95,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            }
        }
    }
}