GPT-5.6 Terra vs MiMo V2.6 Pro: Benchmark Comparison
GPT-5.6 Terra has the higher score on 7 of 18 shared benchmarks; MiMo V2.6 Pro leads on 10.
The largest observed score gap is 48.28 pts on IOI , where GPT-5.6 Terra leads.
Reported ±1 standard-error ranges overlap on 10 of 18 shared benchmarks with comparable uncertainty data. This is not a pairwise statistical significance test.
Shared benchmark results
Scores and reported standard errors come from the latest visible Vals benchmark version available for both models.
| Benchmark | GPT-5.6 Terra | MiMo V2.6 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 53.09% ±1.29 | 55.20% ±1.18 | 2.11 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 0.83% ±0.83 | 10.83% ±2.45 | 10.00 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 41.35% ±3.42 | 47.12% ±3.47 | 5.77 pts | Reported ±1 SE ranges overlap |
| EMB | 66.20% ±3.11 | 62.86% ±3.14 | 3.35 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 54.44% ±2.07 | 57.34% ±0.57 | 2.90 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 65.20% ±3.19 | 64.94% ±3.21 | 0.26 pts | Reported ±1 SE ranges overlap |
| MedCode | 43.41% ±2.17 | 44.97% ±2.10 | 1.55 pts | Reported ±1 SE ranges overlap |
| MedScribe | 82.87% ±1.95 | 88.31% ±1.94 | 5.44 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 74.00% ±4.41 | 70.00% ±4.61 | 4.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench Science | 10.00% ±3.61 | 2.86% ±2.01 | 7.14 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 47.00% ±3.40 | 45.05% ±3.40 | 1.95 pts | Reported ±1 SE ranges overlap |
| Code Migration | 47.80% ±4.28 | 43.01% ±4.32 | 4.79 pts | Reported ±1 SE ranges overlap |
| IOI | 87.61% ±6.53 | 39.33% ±2.41 | 48.28 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.50% ±0.50 | 0.50% ±0.50 | 0.00 pts | Reported ±1 SE ranges overlap |
| Terminal-Bench 4.0 | 22.73% ±2.31 | 31.31% ±3.07 | 8.59 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 74.59% ±4.20 | 85.22% ±3.39 | 10.63 pts | Reported ±1 SE ranges do not overlap |
| CyberBench v1.1 | 72.08% ±5.41 | 72.86% ±5.50 | 0.77 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 62.38% ±1.26 | 68.94% ±1.20 | 6.56 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | GPT-5.6 Terra average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 21.09% | 28.97% |
| Finance | 61.95% | 61.71% |
| Healthcare | 63.14% | 66.64% |
| Math | 74.00% | 70.00% |
| Science | 10.00% | 2.86% |
| Education | 47.00% | 45.05% |
| Coding | 46.65% | 39.88% |
| Cyber | 72.08% | 72.86% |
| Social Mobility | 62.38% | 68.94% |
Cost and latency
Cost per test appears only where the benchmark reports it for both models. Latency is the measured completion time for that benchmark.
| Benchmark | GPT-5.6 Terra cost | MiMo V2.6 Pro cost | GPT-5.6 Terra latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Index | $5.78 | $0.41 | 35m39s | 1h07m |
| Harvey's Legal Agent Benchmark | $3.20 | $0.22 | 17m05s | 25m49s |
| Legal Research Bench | $7.91 | $0.18 | 1h10m | 30m16s |
| EMB | $2.19 | $0.33 | 13m33s | 55m14s |
| Finance Agent (v2) | $3.63 | $0.20 | 24m10s | 10m48s |
| Tax Agent Bench | $4.83 | $0.13 | 1h01m | 25m48s |
| MedCode | N/A | N/A | 18.41s | 2m57s |
| MedScribe | N/A | N/A | 35.53s | 2m53s |
| ProofBench v1.1 | $1.27 | $0.25 | 12m57s | 1h02m |
| Terminal-Bench Science | $5.19 | $0.62 | 2h04m | 3h12m |
| SAGE | N/A | N/A | 40.06s | 3m40s |
| Code Migration | $8.13 | $0.69 | 48m34s | 2h49m |
| IOI | $8.66 | $0.84 | 1h23m | 2h26m |
| ProgramBench | $5.38 | $0.73 | 33m36s | 3h00m |
| Terminal-Bench 4.0 | $5.60 | $0.50 | 31m29s | 2h42m |
| Vibe Code Bench v1.1 | $7.89 | $1.04 | 19m12s | 1h01m |
| CyberBench v1.1 | $3.31 | $0.09 | 17m04s | 27m41s |
| Public Benefits Bench v1.1 | $1.20 | $0.08 | 19m31s | 37m14s |
Results available only for GPT-5.6 Terra
- LegalBench
- MortgageTax
- TaxEval v2
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SkillsBench
- SWE-bench
- Vibe Code Bench 1-100
Results available only for MiMo V2.6 Pro
- MysteryMechanism
- SRE Bench