{
    "metadata": {
        "benchmark": "ProofBench v1.1",
        "slug": "proof_bench",
        "description": "Automated theorem proving benchmark",
        "benchmark_id": "proof_bench",
        "family": "proof_bench",
        "version": "1.1",
        "updated": "2026-10-29",
        "dataset_type": "private",
        "industry": "math",
        "tasks": {
            "overall": "Overall"
        },
        "models": [
            "alibaba/qwen3.8-27b",
            "alibaba/qwen3.8-max",
            "anthropic/claude-fable-5",
            "anthropic/claude-fable-5-1",
            "anthropic/claude-haiku-5-5",
            "anthropic/claude-opus-5",
            "anthropic/claude-opus-5-5",
            "anthropic/claude-sonnet-5",
            "anthropic/claude-sonnet-5-5",
            "aristotle/aristotle",
            "deepseek/deepseek-v4-flash-0731",
            "deepseek/deepseek-v4-pro",
            "deepseek/deepseek-v4-pro-0813",
            "deepseek/deepseek-v4.1-flash",
            "fireworks/ember-1",
            "fireworks/nemotron-lightning-3p5-30b-a3b",
            "google/gemini-3.1-pro-preview",
            "google/gemini-3.5-flash",
            "google/gemini-3.7-flash",
            "google/gemini-3.8-flash",
            "google/gemini-4-argon",
            "grok/grok-4.5",
            "grok/grok-4.6",
            "grok/grok-4.7",
            "inception/mercury-2.5",
            "kimi/kimi-k3",
            "kimi/kimi-k3.1-pro",
            "logicalintelligence/alephprover",
            "meta/muse-wonder",
            "meta/muse-wonder-full",
            "meta/muse_spark_1_2",
            "meta/muse_spark_1_3",
            "meta/muse_spark_1_3_max",
            "minimax/MiniMax-M3",
            "mistralai/mistral-large-4",
            "mistralai/mistral-medium-3.5",
            "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",
            "reflection/Beam-1.0",
            "stepfun/step-5-preview",
            "tencent/hy4-preview",
            "thinkingmachines/inkling",
            "thinkingmachines/inkling-small",
            "xiaomi/mimo-v2.5",
            "xiaomi/mimo-v2.5-pro",
            "xiaomi/mimo-v2.6-flash",
            "xiaomi/mimo-v2.6-pro",
            "zai/glm-5.3",
            "zai/glm-5.3-flash"
        ],
        "partners": [],
        "showBadge": false,
        "visible": true,
        "use_cost_per_test": true,
        "token_tasks_measured": 100,
        "runner": "external",
        "mode": "agentic",
        "archived": false,
        "partner": false,
        "total_models": 54
    },
    "tasks": {
        "overall": {
            "anthropic/claude-sonnet-5-5": {
                "accuracy": 100.0,
                "latency": 347.583,
                "stderr": 0.0,
                "cost_per_test": 0.558327,
                "token_totals": {
                    "input_tokens": 38210470,
                    "output_tokens": 3817407,
                    "reasoning_tokens": 2302419,
                    "cache_read_tokens": 33854803,
                    "cache_write_tokens": 4352571
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 7.7,
                    "avg_query_seconds": 44.9,
                    "median_query_seconds": 33.3,
                    "avg_ttft_seconds": 27.1,
                    "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-opus-5-5": {
                "accuracy": 100.0,
                "latency": 351.035,
                "stderr": 0.0,
                "cost_per_test": 0.961648,
                "token_totals": {
                    "input_tokens": 22741291,
                    "output_tokens": 3445356,
                    "reasoning_tokens": 2381878,
                    "cache_read_tokens": 18792754,
                    "cache_write_tokens": 3946449
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 5.2,
                    "avg_query_seconds": 67.2,
                    "median_query_seconds": 36.5,
                    "avg_ttft_seconds": 53.3,
                    "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": 100.0,
                "latency": 478.303,
                "stderr": 0.0,
                "cost_per_test": 2.70347,
                "token_totals": {
                    "input_tokens": 25650116,
                    "output_tokens": 4101355,
                    "reasoning_tokens": 2989018,
                    "cache_read_tokens": 20844282,
                    "cache_write_tokens": 4803934
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 4.8,
                    "avg_query_seconds": 100.7,
                    "median_query_seconds": 50.8,
                    "avg_ttft_seconds": 74.1,
                    "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
            },
            "logicalintelligence/alephprover": {
                "accuracy": 100.0,
                "latency": 1108.1,
                "stderr": 0.0,
                "cost_per_test": 9.3487,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Logical Intelligence",
                "harness": null
            },
            "openai/gpt-6.1-sol": {
                "accuracy": 99.0,
                "latency": 284.697,
                "stderr": 1.0,
                "cost_per_test": 0.233057,
                "token_totals": {
                    "input_tokens": 32842768,
                    "output_tokens": 1038054,
                    "reasoning_tokens": 321606,
                    "cache_read_tokens": 28854579,
                    "cache_write_tokens": 3985465
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 9.1,
                    "avg_query_seconds": 31.3,
                    "median_query_seconds": 25.7,
                    "avg_ttft_seconds": 12.0,
                    "output_time_split": {
                        "reasoning_pct": 11.6,
                        "content_pct": 0.0,
                        "tool_call_pct": 88.4
                    }
                },
                "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": 99.0,
                "latency": 307.867,
                "stderr": 1.0,
                "cost_per_test": 1.71238,
                "token_totals": {
                    "input_tokens": 35641964,
                    "output_tokens": 997178,
                    "reasoning_tokens": 309366,
                    "cache_read_tokens": 29434576,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 8.7,
                    "avg_query_seconds": 35.3,
                    "median_query_seconds": 30.1,
                    "avg_ttft_seconds": 8.8,
                    "output_time_split": {
                        "reasoning_pct": 11.8,
                        "content_pct": 0.0,
                        "tool_call_pct": 88.2
                    }
                },
                "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-argon": {
                "accuracy": 99.0,
                "latency": 380.246,
                "stderr": 1.0,
                "cost_per_test": 0.782978,
                "token_totals": {
                    "input_tokens": 37238434,
                    "output_tokens": 1694278,
                    "reasoning_tokens": 702938,
                    "cache_read_tokens": 27510930,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 10.8,
                    "avg_query_seconds": 35.3,
                    "median_query_seconds": 27.7,
                    "avg_ttft_seconds": 6.8,
                    "output_time_split": {
                        "reasoning_pct": 92.4,
                        "content_pct": 0.0,
                        "tool_call_pct": 7.6
                    }
                },
                "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": 99.0,
                "latency": 590.886,
                "stderr": 1.0,
                "cost_per_test": 1.786125,
                "token_totals": {
                    "input_tokens": 51018430,
                    "output_tokens": 4769719,
                    "reasoning_tokens": 3301232,
                    "cache_read_tokens": 45129298,
                    "cache_write_tokens": 5887376
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 8.8,
                    "avg_query_seconds": 67.3,
                    "median_query_seconds": 42.9,
                    "avg_ttft_seconds": 46.7,
                    "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": 95.0,
                "latency": 822.739,
                "stderr": 2.19,
                "cost_per_test": 5.229282,
                "token_totals": {
                    "input_tokens": 35941254,
                    "output_tokens": 7112728,
                    "reasoning_tokens": 5678640,
                    "cache_read_tokens": 24973984,
                    "cache_write_tokens": 10966348
                },
                "token_tasks_measured": 99,
                "turn_metrics": {
                    "tasks_measured": 99,
                    "avg_turns": 4.7,
                    "avg_query_seconds": 176.7,
                    "median_query_seconds": 102.4,
                    "avg_ttft_seconds": 150.4,
                    "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
            },
            "meta/muse-wonder-full": {
                "accuracy": 94.0,
                "latency": 695.558,
                "stderr": 2.387,
                "cost_per_test": 0.856208,
                "token_totals": {
                    "input_tokens": 62718296,
                    "output_tokens": 2708964,
                    "reasoning_tokens": 1349584,
                    "cache_read_tokens": 57745168,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 17.7,
                    "avg_query_seconds": 39.2,
                    "median_query_seconds": 26.8,
                    "avg_ttft_seconds": 21.6,
                    "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
            },
            "kimi/kimi-k3.1-pro": {
                "accuracy": 91.0,
                "latency": 496.158,
                "stderr": 2.876,
                "cost_per_test": 0.474341,
                "token_totals": {
                    "input_tokens": 73516483,
                    "output_tokens": 2238130,
                    "reasoning_tokens": 1202518,
                    "cache_read_tokens": 67766784,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 15.0,
                    "avg_query_seconds": 33.0,
                    "median_query_seconds": 18.3,
                    "avg_ttft_seconds": 1.4,
                    "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
            },
            "anthropic/claude-haiku-5-5": {
                "accuracy": 89.0,
                "latency": 418.047,
                "stderr": 3.145,
                "cost_per_test": 0.173874,
                "token_totals": {
                    "input_tokens": 42301619,
                    "output_tokens": 9672099,
                    "reasoning_tokens": 8496068,
                    "cache_read_tokens": 29799415,
                    "cache_write_tokens": 12500492
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 4.3,
                    "avg_query_seconds": 96.5,
                    "median_query_seconds": 50.3,
                    "avg_ttft_seconds": 73.4,
                    "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
            },
            "kimi/kimi-k3": {
                "accuracy": 87.0,
                "latency": 2070.511,
                "stderr": 3.38,
                "cost_per_test": 1.646802,
                "token_totals": {
                    "input_tokens": 104584764,
                    "output_tokens": 7319059,
                    "reasoning_tokens": 5780416,
                    "cache_read_tokens": 95874048,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 13.4,
                    "avg_query_seconds": 154.0,
                    "median_query_seconds": 76.0,
                    "avg_ttft_seconds": 14.3,
                    "output_time_split": {
                        "reasoning_pct": 79.9,
                        "content_pct": 0.6,
                        "tool_call_pct": 19.5
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 262144,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Moonshot AI",
                "harness": null
            },
            "aristotle/aristotle": {
                "accuracy": 86.0,
                "latency": null,
                "stderr": 3.47,
                "cost_per_test": null,
                "temperature": null,
                "top_p": null,
                "max_output_tokens": null,
                "reasoning": null,
                "reasoning_effort": null,
                "verbosity": null,
                "compute_effort": null,
                "provider": "Harmonic",
                "harness": null
            },
            "meta/muse-wonder": {
                "accuracy": 84.0,
                "latency": 929.053,
                "stderr": 3.685,
                "cost_per_test": 1.313345,
                "token_totals": {
                    "input_tokens": 83508776,
                    "output_tokens": 3687574,
                    "reasoning_tokens": 2180479,
                    "cache_read_tokens": 72750544,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 21.0,
                    "avg_query_seconds": 44.1,
                    "median_query_seconds": 26.0,
                    "avg_ttft_seconds": 29.9,
                    "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
            },
            "openai/gpt-6-sol": {
                "accuracy": 83.0,
                "latency": 304.509,
                "stderr": 3.775,
                "cost_per_test": 0.476321,
                "token_totals": {
                    "input_tokens": 99039756,
                    "output_tokens": 1363035,
                    "reasoning_tokens": 593322,
                    "cache_read_tokens": 92867287,
                    "cache_write_tokens": 6166643
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 19.4,
                    "avg_query_seconds": 15.7,
                    "median_query_seconds": 13.1,
                    "avg_ttft_seconds": 5.4,
                    "output_time_split": {
                        "reasoning_pct": 15.3,
                        "content_pct": 0.0,
                        "tool_call_pct": 84.7
                    }
                },
                "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": 83.0,
                "latency": 520.405,
                "stderr": 3.775,
                "cost_per_test": 1.438313,
                "token_totals": {
                    "input_tokens": 123160663,
                    "output_tokens": 1522264,
                    "reasoning_tokens": 541932,
                    "cache_read_tokens": 115350727,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 99,
                "turn_metrics": {
                    "tasks_measured": 99,
                    "avg_turns": 18.0,
                    "avg_query_seconds": 28.9,
                    "median_query_seconds": 20.7,
                    "avg_ttft_seconds": 10.1,
                    "output_time_split": {
                        "reasoning_pct": 22.2,
                        "content_pct": 0.0,
                        "tool_call_pct": 77.8
                    }
                },
                "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": 77.0,
                "latency": 898.079,
                "stderr": 4.23,
                "cost_per_test": 1.373684,
                "token_totals": {
                    "input_tokens": 158153304,
                    "output_tokens": 7432445,
                    "reasoning_tokens": 5551345,
                    "cache_read_tokens": 145688844,
                    "cache_write_tokens": 12459810
                },
                "token_tasks_measured": 98,
                "turn_metrics": {
                    "tasks_measured": 98,
                    "avg_turns": 23.7,
                    "avg_query_seconds": 37.8,
                    "median_query_seconds": 23.8,
                    "avg_ttft_seconds": 27.6,
                    "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
            },
            "tencent/hy4-preview": {
                "accuracy": 75.0,
                "latency": 1373.837,
                "stderr": 4.352,
                "cost_per_test": 0.335717,
                "token_totals": {
                    "input_tokens": 134713193,
                    "output_tokens": 7030036,
                    "reasoning_tokens": 4997323,
                    "cache_read_tokens": 121668224,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 19.1,
                    "avg_query_seconds": 71.9,
                    "median_query_seconds": 50.3,
                    "avg_ttft_seconds": 3.4,
                    "output_time_split": {
                        "reasoning_pct": 77.7,
                        "content_pct": 0.1,
                        "tool_call_pct": 22.2
                    }
                },
                "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": 74.0,
                "latency": 777.341,
                "stderr": 4.408,
                "cost_per_test": 1.274209,
                "token_totals": {
                    "input_tokens": 241131910,
                    "output_tokens": 1542468,
                    "reasoning_tokens": 704967,
                    "cache_read_tokens": 222728984,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 99,
                "turn_metrics": {
                    "tasks_measured": 99,
                    "avg_turns": 21.4,
                    "avg_query_seconds": 36.3,
                    "median_query_seconds": 21.6,
                    "avg_ttft_seconds": 5.4,
                    "output_time_split": {
                        "reasoning_pct": 29.1,
                        "content_pct": 0.0,
                        "tool_call_pct": 70.9
                    }
                },
                "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": 70.0,
                "latency": 3693.469,
                "stderr": 4.606,
                "cost_per_test": 0.24535,
                "token_totals": {
                    "input_tokens": 214175878,
                    "output_tokens": 15653193,
                    "reasoning_tokens": 13784976,
                    "cache_read_tokens": 192932736,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 13.5,
                    "avg_query_seconds": 273.6,
                    "median_query_seconds": 152.1,
                    "avg_ttft_seconds": 10.4,
                    "output_time_split": {
                        "reasoning_pct": 92.5,
                        "content_pct": 0.2,
                        "tool_call_pct": 7.3
                    }
                },
                "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
            },
            "openai/gpt-6-luna": {
                "accuracy": 64.0,
                "latency": 451.481,
                "stderr": 4.824,
                "cost_per_test": 0.041637,
                "token_totals": {
                    "input_tokens": 126053768,
                    "output_tokens": 2652752,
                    "reasoning_tokens": 1912396,
                    "cache_read_tokens": 117050095,
                    "cache_write_tokens": 8996692
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 23.3,
                    "avg_query_seconds": 19.4,
                    "median_query_seconds": 13.8,
                    "avg_ttft_seconds": 7.4,
                    "output_time_split": {
                        "reasoning_pct": 31.2,
                        "content_pct": 0.0,
                        "tool_call_pct": 68.8
                    }
                },
                "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-flash": {
                "accuracy": 63.0,
                "latency": 3769.808,
                "stderr": 4.852,
                "cost_per_test": 0.123277,
                "token_totals": {
                    "input_tokens": 398251867,
                    "output_tokens": 24214511,
                    "reasoning_tokens": 20726471,
                    "cache_read_tokens": 365944512,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 17.0,
                    "avg_query_seconds": 222.0,
                    "median_query_seconds": 143.6,
                    "avg_ttft_seconds": 8.4,
                    "output_time_split": {
                        "reasoning_pct": 20.4,
                        "content_pct": 0.0,
                        "tool_call_pct": 79.6
                    }
                },
                "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
            },
            "fireworks/ember-1": {
                "accuracy": 61.0,
                "latency": 1269.194,
                "stderr": 4.902,
                "cost_per_test": 1.737739,
                "token_totals": {
                    "input_tokens": 111831127,
                    "output_tokens": 7247825,
                    "reasoning_tokens": 5344776,
                    "cache_read_tokens": 100161789,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 15.3,
                    "avg_query_seconds": 82.9,
                    "median_query_seconds": 37.5,
                    "avg_ttft_seconds": 2.3,
                    "output_time_split": {
                        "reasoning_pct": 73.5,
                        "content_pct": 0.4,
                        "tool_call_pct": 26.1
                    }
                },
                "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-luna": {
                "accuracy": 60.0,
                "latency": 513.687,
                "stderr": 4.924,
                "cost_per_test": 0.123288,
                "token_totals": {
                    "input_tokens": 203002670,
                    "output_tokens": 1885904,
                    "reasoning_tokens": 1035104,
                    "cache_read_tokens": 184422820,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 97,
                "turn_metrics": {
                    "tasks_measured": 97,
                    "avg_turns": 25.1,
                    "avg_query_seconds": 20.5,
                    "median_query_seconds": 15.9,
                    "avg_ttft_seconds": 6.8,
                    "output_time_split": {
                        "reasoning_pct": 30.7,
                        "content_pct": 0.0,
                        "tool_call_pct": 69.3
                    }
                },
                "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": 58.0,
                "latency": 423.698,
                "stderr": 4.96,
                "cost_per_test": 0.558508,
                "token_totals": {
                    "input_tokens": 58411119,
                    "output_tokens": 1397278,
                    "reasoning_tokens": 567660,
                    "cache_read_tokens": 31292904,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 29.9,
                    "avg_query_seconds": 14.2,
                    "median_query_seconds": 10.4,
                    "avg_ttft_seconds": 2.9,
                    "output_time_split": {
                        "reasoning_pct": 91.1,
                        "content_pct": 0.0,
                        "tool_call_pct": 8.9
                    }
                },
                "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": 58.0,
                "latency": 568.696,
                "stderr": 4.96,
                "cost_per_test": 0.479019,
                "token_totals": {
                    "input_tokens": 110539459,
                    "output_tokens": 4529213,
                    "reasoning_tokens": 3431454,
                    "cache_read_tokens": 99565051,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 30.1,
                    "avg_query_seconds": 18.9,
                    "median_query_seconds": 10.8,
                    "avg_ttft_seconds": 4.0,
                    "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
            },
            "alibaba/qwen3.8-max": {
                "accuracy": 58.0,
                "latency": 1867.622,
                "stderr": 4.96,
                "cost_per_test": 1.503984,
                "token_totals": {
                    "input_tokens": 207535531,
                    "output_tokens": 10505110,
                    "reasoning_tokens": 9299092,
                    "cache_read_tokens": 190696704,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 96,
                "turn_metrics": {
                    "tasks_measured": 96,
                    "avg_turns": 27.1,
                    "avg_query_seconds": 69.0,
                    "median_query_seconds": 22.2,
                    "avg_ttft_seconds": 7.2,
                    "output_time_split": {
                        "reasoning_pct": 87.6,
                        "content_pct": 0.1,
                        "tool_call_pct": 12.2
                    }
                },
                "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-flash-0731": {
                "accuracy": 56.0,
                "latency": 933.981,
                "stderr": 4.989,
                "cost_per_test": 0.156045,
                "token_totals": {
                    "input_tokens": 168698684,
                    "output_tokens": 7445555,
                    "reasoning_tokens": 5646846,
                    "cache_read_tokens": 160683264,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 27.7,
                    "avg_query_seconds": 33.7,
                    "median_query_seconds": 22.3,
                    "avg_ttft_seconds": 1.4,
                    "output_time_split": {
                        "reasoning_pct": 82.8,
                        "content_pct": 0.0,
                        "tool_call_pct": 17.1
                    }
                },
                "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_3": {
                "accuracy": 55.0,
                "latency": 534.06,
                "stderr": 5.0,
                "cost_per_test": 0.433712,
                "token_totals": {
                    "input_tokens": 106926796,
                    "output_tokens": 3606435,
                    "reasoning_tokens": 2473887,
                    "cache_read_tokens": 96013349,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 29.7,
                    "avg_query_seconds": 17.9,
                    "median_query_seconds": 11.7,
                    "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-v4.1-flash": {
                "accuracy": 54.0,
                "latency": 546.595,
                "stderr": 5.009,
                "cost_per_test": 0.126437,
                "token_totals": {
                    "input_tokens": 183256713,
                    "output_tokens": 8190207,
                    "reasoning_tokens": 6720807,
                    "cache_read_tokens": 177420288,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 20.2,
                    "avg_query_seconds": 27.1,
                    "median_query_seconds": 20.1,
                    "avg_ttft_seconds": 1.6,
                    "output_time_split": {
                        "reasoning_pct": 84.7,
                        "content_pct": 0.2,
                        "tool_call_pct": 15.1
                    }
                },
                "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-4.6": {
                "accuracy": 51.0,
                "latency": 861.898,
                "stderr": 5.024,
                "cost_per_test": 0.824475,
                "token_totals": {
                    "input_tokens": 86079791,
                    "output_tokens": 3827061,
                    "reasoning_tokens": 2540626,
                    "cache_read_tokens": 75116288,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 29.4,
                    "avg_query_seconds": 29.3,
                    "median_query_seconds": 17.0,
                    "avg_ttft_seconds": 1.5,
                    "output_time_split": {
                        "reasoning_pct": 73.7,
                        "content_pct": 24.7,
                        "tool_call_pct": 1.6
                    }
                },
                "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-pro-0813": {
                "accuracy": 50.0,
                "latency": 732.365,
                "stderr": 5.025,
                "cost_per_test": 0.262008,
                "token_totals": {
                    "input_tokens": 133104897,
                    "output_tokens": 2434507,
                    "reasoning_tokens": 1527005,
                    "cache_read_tokens": 124716544,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 29.5,
                    "avg_query_seconds": 24.8,
                    "median_query_seconds": 14.5,
                    "avg_ttft_seconds": 1.6,
                    "output_time_split": {
                        "reasoning_pct": 73.5,
                        "content_pct": 0.0,
                        "tool_call_pct": 26.5
                    }
                },
                "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-5.3": {
                "accuracy": 49.0,
                "latency": 2542.672,
                "stderr": 5.024,
                "cost_per_test": 2.081538,
                "token_totals": {
                    "input_tokens": 275938147,
                    "output_tokens": 19360716,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 232831819,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 99,
                "turn_metrics": {
                    "tasks_measured": 99,
                    "avg_turns": 14.7,
                    "avg_query_seconds": 172.5,
                    "median_query_seconds": 117.7,
                    "avg_ttft_seconds": 4.6,
                    "output_time_split": {
                        "reasoning_pct": 89.9,
                        "content_pct": 0.3,
                        "tool_call_pct": 9.8
                    }
                },
                "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": 48.0,
                "latency": 368.648,
                "stderr": 5.021,
                "cost_per_test": 0.602872,
                "token_totals": {
                    "input_tokens": 69594721,
                    "output_tokens": 1221773,
                    "reasoning_tokens": 738421,
                    "cache_read_tokens": 40351064,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 98,
                "turn_metrics": {
                    "tasks_measured": 98,
                    "avg_turns": 32.5,
                    "avg_query_seconds": 11.3,
                    "median_query_seconds": 9.0,
                    "avg_ttft_seconds": 1.6,
                    "output_time_split": {
                        "reasoning_pct": 95.5,
                        "content_pct": 0.0,
                        "tool_call_pct": 4.5
                    }
                },
                "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": 43.0,
                "latency": 469.146,
                "stderr": 4.976,
                "cost_per_test": 0.442007,
                "token_totals": {
                    "input_tokens": 104594507,
                    "output_tokens": 3716809,
                    "reasoning_tokens": 2515620,
                    "cache_read_tokens": 94642678,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 96,
                "turn_metrics": {
                    "tasks_measured": 96,
                    "avg_turns": 32.5,
                    "avg_query_seconds": 14.4,
                    "median_query_seconds": 10.3,
                    "avg_ttft_seconds": 7.1,
                    "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
            },
            "stepfun/step-5-preview": {
                "accuracy": 42.0,
                "latency": 2313.592,
                "stderr": 4.96,
                "cost_per_test": 0.714564,
                "token_totals": {
                    "input_tokens": 258573571,
                    "output_tokens": 13318410,
                    "reasoning_tokens": 0,
                    "cache_read_tokens": 234817792,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 31.1,
                    "avg_query_seconds": 74.3,
                    "median_query_seconds": 34.6,
                    "avg_ttft_seconds": 4.0,
                    "output_time_split": {
                        "reasoning_pct": 89.1,
                        "content_pct": 1.0,
                        "tool_call_pct": 9.8
                    }
                },
                "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-3.5-flash": {
                "accuracy": 31.0,
                "latency": 394.476,
                "stderr": 4.648,
                "cost_per_test": 0.569863,
                "token_totals": {
                    "input_tokens": 66676489,
                    "output_tokens": 1338730,
                    "reasoning_tokens": 805865,
                    "cache_read_tokens": 42908380,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 95,
                "turn_metrics": {
                    "tasks_measured": 95,
                    "avg_turns": 34.4,
                    "avg_query_seconds": 11.5,
                    "median_query_seconds": 9.5,
                    "avg_ttft_seconds": 2.3,
                    "output_time_split": {
                        "reasoning_pct": 89.1,
                        "content_pct": 9.2,
                        "tool_call_pct": 1.7
                    }
                },
                "temperature": 1.0,
                "top_p": null,
                "max_output_tokens": 65536,
                "reasoning": null,
                "reasoning_effort": "high",
                "verbosity": null,
                "compute_effort": null,
                "provider": "Google",
                "harness": null
            },
            "grok/grok-4.5": {
                "accuracy": 31.0,
                "latency": 668.86,
                "stderr": 4.648,
                "cost_per_test": 0.573991,
                "token_totals": {
                    "input_tokens": 74344565,
                    "output_tokens": 2301329,
                    "reasoning_tokens": 606173,
                    "cache_read_tokens": 64634880,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 29.1,
                    "avg_query_seconds": 23.0,
                    "median_query_seconds": 16.6,
                    "avg_ttft_seconds": 1.5,
                    "output_time_split": {
                        "reasoning_pct": 41.1,
                        "content_pct": 55.3,
                        "tool_call_pct": 3.6
                    }
                },
                "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
            },
            "grok/grok-4.7": {
                "accuracy": 26.0,
                "latency": 767.177,
                "stderr": 4.408,
                "cost_per_test": 0.787951,
                "token_totals": {
                    "input_tokens": 70796678,
                    "output_tokens": 1422566,
                    "reasoning_tokens": 829928,
                    "cache_read_tokens": 50374400,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 35.9,
                    "avg_query_seconds": 21.3,
                    "median_query_seconds": 12.5,
                    "avg_ttft_seconds": 2.9,
                    "output_time_split": {
                        "reasoning_pct": 94.7,
                        "content_pct": 0.7,
                        "tool_call_pct": 4.7
                    }
                },
                "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-3.1-pro-preview": {
                "accuracy": 26.0,
                "latency": 889.059,
                "stderr": 4.408,
                "cost_per_test": 0.784253,
                "token_totals": {
                    "input_tokens": 68260388,
                    "output_tokens": 2100047,
                    "reasoning_tokens": 691664,
                    "cache_read_tokens": 47066256,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 29.9,
                    "avg_query_seconds": 29.7,
                    "median_query_seconds": 19.8,
                    "avg_ttft_seconds": 11.9,
                    "output_time_split": {
                        "reasoning_pct": 56.8,
                        "content_pct": 4.4,
                        "tool_call_pct": 38.8
                    }
                },
                "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": 22.0,
                "latency": 1177.352,
                "stderr": 4.163,
                "cost_per_test": 0.089853,
                "token_totals": {
                    "input_tokens": 109400066,
                    "output_tokens": 5301519,
                    "reasoning_tokens": 0,
                    "cache_read_tokens": 100176384,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 34.2,
                    "avg_query_seconds": 34.4,
                    "median_query_seconds": 17.2,
                    "avg_ttft_seconds": 3.3,
                    "output_time_split": {
                        "reasoning_pct": 82.6,
                        "content_pct": 0.4,
                        "tool_call_pct": 17.0
                    }
                },
                "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
            },
            "zai/glm-5.3-flash": {
                "accuracy": 21.0,
                "latency": 4274.803,
                "stderr": 4.094,
                "cost_per_test": 0.224234,
                "token_totals": {
                    "input_tokens": 703145622,
                    "output_tokens": 33321229,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 644046720,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 23.2,
                    "avg_query_seconds": 184.2,
                    "median_query_seconds": 74.2,
                    "avg_ttft_seconds": 3.2,
                    "output_time_split": {
                        "reasoning_pct": 92.4,
                        "content_pct": 1.6,
                        "tool_call_pct": 5.9
                    }
                },
                "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
            },
            "reflection/Beam-1.0": {
                "accuracy": 18.0,
                "latency": 266.658,
                "stderr": 3.861,
                "cost_per_test": null,
                "token_totals": {
                    "input_tokens": 64179687,
                    "output_tokens": 1784980,
                    "reasoning_tokens": 1415418,
                    "cache_read_tokens": 56986688,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 36.7,
                    "avg_query_seconds": 7.2,
                    "median_query_seconds": 3.5,
                    "avg_ttft_seconds": 1.6,
                    "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
            },
            "minimax/MiniMax-M3": {
                "accuracy": 18.0,
                "latency": 718.66,
                "stderr": 3.861,
                "cost_per_test": 0.419224,
                "token_totals": {
                    "input_tokens": 162739395,
                    "output_tokens": 4876274,
                    "reasoning_tokens": 1498488,
                    "cache_read_tokens": 151053885,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 33.8,
                    "avg_query_seconds": 21.2,
                    "median_query_seconds": 11.0,
                    "avg_ttft_seconds": 2.0,
                    "output_time_split": {
                        "reasoning_pct": 87.8,
                        "content_pct": 1.6,
                        "tool_call_pct": 10.6
                    }
                },
                "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": 16.0,
                "latency": 423.45,
                "stderr": 3.685,
                "cost_per_test": 0.142045,
                "token_totals": {
                    "input_tokens": 61091499,
                    "output_tokens": 1611579,
                    "reasoning_tokens": 1200572,
                    "cache_read_tokens": 57067520,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 37.5,
                    "avg_query_seconds": 11.3,
                    "median_query_seconds": 7.2,
                    "avg_ttft_seconds": 1.4,
                    "output_time_split": {
                        "reasoning_pct": 85.7,
                        "content_pct": 0.0,
                        "tool_call_pct": 14.3
                    }
                },
                "temperature": null,
                "top_p": null,
                "max_output_tokens": 128000,
                "reasoning": null,
                "reasoning_effort": "max",
                "verbosity": null,
                "compute_effort": null,
                "provider": "DeepSeek",
                "harness": null
            },
            "xiaomi/mimo-v2.5": {
                "accuracy": 16.0,
                "latency": 1255.813,
                "stderr": 3.685,
                "cost_per_test": 0.05707,
                "token_totals": {
                    "input_tokens": 139624413,
                    "output_tokens": 11716945,
                    "reasoning_tokens": 10908282,
                    "cache_read_tokens": 125621632,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 98,
                "turn_metrics": {
                    "tasks_measured": 98,
                    "avg_turns": 33.1,
                    "avg_query_seconds": 38.0,
                    "median_query_seconds": 14.8,
                    "avg_ttft_seconds": 3.3,
                    "output_time_split": {
                        "reasoning_pct": 92.1,
                        "content_pct": 0.3,
                        "tool_call_pct": 7.6
                    }
                },
                "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
            },
            "alibaba/qwen3.8-27b": {
                "accuracy": 16.0,
                "latency": 1453.872,
                "stderr": 3.685,
                "cost_per_test": 2.298832,
                "token_totals": {
                    "input_tokens": 358351531,
                    "output_tokens": 16902495,
                    "reasoning_tokens": 14195905,
                    "cache_read_tokens": 0,
                    "cache_write_tokens": null
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 30.0,
                    "avg_query_seconds": 48.5,
                    "median_query_seconds": 24.2,
                    "avg_ttft_seconds": 4.8,
                    "output_time_split": {
                        "reasoning_pct": 84.2,
                        "content_pct": 0.4,
                        "tool_call_pct": 15.4
                    }
                },
                "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": 10.0,
                "latency": 1258.806,
                "stderr": 3.015,
                "cost_per_test": 0.329279,
                "token_totals": {
                    "input_tokens": 59543255,
                    "output_tokens": 1823695,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 19971584,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 36.9,
                    "avg_query_seconds": 34.1,
                    "median_query_seconds": 21.1,
                    "avg_ttft_seconds": 5.1,
                    "output_time_split": {
                        "reasoning_pct": 99.8,
                        "content_pct": 0.0,
                        "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
            },
            "mistralai/mistral-medium-3.5": {
                "accuracy": 9.0,
                "latency": 479.048,
                "stderr": 2.876,
                "cost_per_test": 1.302032,
                "token_totals": {
                    "input_tokens": 66324649,
                    "output_tokens": 1838640,
                    "reasoning_tokens": null,
                    "cache_read_tokens": null,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 87,
                "turn_metrics": {
                    "tasks_measured": 87,
                    "avg_turns": 36.5,
                    "avg_query_seconds": 13.1,
                    "median_query_seconds": 7.8,
                    "avg_ttft_seconds": 4.5,
                    "output_time_split": {
                        "reasoning_pct": 96.2,
                        "content_pct": 3.2,
                        "tool_call_pct": 0.5
                    }
                },
                "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
            },
            "thinkingmachines/inkling-small": {
                "accuracy": 6.0,
                "latency": 507.301,
                "stderr": 2.387,
                "cost_per_test": 0.082234,
                "token_totals": {
                    "input_tokens": 62832266,
                    "output_tokens": 2596974,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 57261056,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 37.7,
                    "avg_query_seconds": 13.4,
                    "median_query_seconds": 8.7,
                    "avg_ttft_seconds": 1.8,
                    "output_time_split": {
                        "reasoning_pct": 99.9,
                        "content_pct": 0.1,
                        "tool_call_pct": 0.1
                    }
                },
                "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
            },
            "inception/mercury-2.5": {
                "accuracy": 3.0,
                "latency": 201.725,
                "stderr": 1.714,
                "cost_per_test": 0.044105,
                "token_totals": {
                    "input_tokens": 54269237,
                    "output_tokens": 1560820,
                    "reasoning_tokens": 1435858,
                    "cache_read_tokens": 42299649,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 33.3,
                    "avg_query_seconds": 6.0,
                    "median_query_seconds": 5.0,
                    "avg_ttft_seconds": 1.4,
                    "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
            },
            "thinkingmachines/inkling": {
                "accuracy": 0.0,
                "latency": 436.43,
                "stderr": 0.0,
                "cost_per_test": 0.198639,
                "token_totals": {
                    "input_tokens": 56824656,
                    "output_tokens": 1417408,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 51447296,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 39.4,
                    "avg_query_seconds": 11.1,
                    "median_query_seconds": 7.5,
                    "avg_ttft_seconds": 1.6,
                    "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
            },
            "fireworks/nemotron-lightning-3p5-30b-a3b": {
                "accuracy": 0.0,
                "latency": 919.248,
                "stderr": 0.0,
                "cost_per_test": 0.015926,
                "token_totals": {
                    "input_tokens": 43944505,
                    "output_tokens": 5131368,
                    "reasoning_tokens": null,
                    "cache_read_tokens": 40772695,
                    "cache_write_tokens": 0
                },
                "token_tasks_measured": 100,
                "turn_metrics": {
                    "tasks_measured": 100,
                    "avg_turns": 36.2,
                    "avg_query_seconds": 25.4,
                    "median_query_seconds": 9.4,
                    "avg_ttft_seconds": 4.0,
                    "output_time_split": {
                        "reasoning_pct": 72.4,
                        "content_pct": 22.2,
                        "tool_call_pct": 5.4
                    }
                },
                "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
            }
        }
    }
}