Gemini 3.1 Pro Preview (02/26) vs MiMo V2.5: Benchmark Comparison
Gemini 3.1 Pro Preview (02/26) has the higher score on 16 of 19 shared benchmarks; MiMo V2.5 leads on 3.
The largest observed score gap is 27.17 pts on MedCode , where Gemini 3.1 Pro Preview (02/26) leads.
Reported ±1 standard-error ranges overlap on 4 of 19 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 | Gemini 3.1 Pro Preview (02/26) | MiMo V2.5 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 0.00% ±0.00 | 1.67% ±0.00 | 1.67 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 20.67% ±2.81 | 9.13% ±2.00 | 11.54 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 87.40% ±0.33 | 78.89% ±0.51 | 8.51 pts | Reported ±1 SE ranges do not overlap |
| EMB | 52.62% ±2.97 | 55.09% ±3.13 | 2.48 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 42.98% ±1.21 | 36.73% ±0.11 | 6.25 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 69.40% ±0.91 | 59.26% ±0.98 | 10.14 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 37.84% ±3.10 | 29.16% ±2.71 | 8.68 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 72.88% ±0.86 | 71.83% ±0.88 | 1.05 pts | Reported ±1 SE ranges overlap |
| MedCode | 59.06% ±2.00 | 31.89% ±2.02 | 27.17 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 76.11% ±1.92 | 72.15% ±1.85 | 3.96 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 26.00% ±4.41 | 16.00% ±3.69 | 10.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 95.45% ±1.05 | 81.57% ±2.05 | 13.89 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 90.99% ±0.28 | 82.93% ±0.37 | 8.06 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 88.21% ±0.78 | 80.00% ±0.96 | 8.21 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 48.68% ±3.29 | 43.27% ±3.39 | 5.41 pts | Reported ±1 SE ranges overlap |
| Code Migration | 17.31% ±3.93 | 14.23% ±3.69 | 3.08 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 88.48% ±0.93 | 81.51% ±1.07 | 6.97 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 78.80% ±1.83 | 71.00% ±2.03 | 7.80 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 32.03% ±4.34 | 42.17% ±4.57 | 10.14 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 3.1 Pro Preview (02/26) average | MiMo V2.5 average |
|---|---|---|
| Legal | 36.02% | 29.90% |
| Finance | 55.14% | 50.41% |
| Healthcare | 67.59% | 52.02% |
| Math | 26.00% | 16.00% |
| Academic | 91.55% | 81.50% |
| Education | 48.68% | 43.27% |
| Coding | 54.16% | 52.23% |
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 | Gemini 3.1 Pro Preview (02/26) cost | MiMo V2.5 cost | Gemini 3.1 Pro Preview (02/26) latency | MiMo V2.5 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $1.15 | $0.04 | 7m17s | 6m59s |
| Legal Research Bench | $1.10 | $0.03 | 4m02s | 6m31s |
| LegalBench | N/A | N/A | 10.06s | 11.18s |
| EMB | $3.77 | $0.08 | 11m49s | 30m38s |
| Finance Agent (v2) | $1.59 | $0.09 | 3m28s | 5m39s |
| MortgageTax | N/A | N/A | 23.62s | 32.94s |
| Tax Agent Bench | $0.30 | $0.02 | 100.96s | 8m21s |
| TaxEval v2 | N/A | N/A | 41.93s | 26.30s |
| MedCode | N/A | N/A | 38.52s | 17.70s |
| MedScribe | N/A | N/A | 69.14s | 20.26s |
| ProofBench v1.1 | $0.78 | $0.06 | 14m49s | 20m56s |
| GPQA Diamond | N/A | N/A | 65.76s | 89.39s |
| MMLU Pro | N/A | N/A | 23.91s | 30.25s |
| MMMU Pro | N/A | N/A | 76.99s | 42.47s |
| SAGE | N/A | N/A | 61.17s | 52.66s |
| Code Migration | $1.61 | $0.10 | 12m01s | 37m55s |
| LiveCodeBench | N/A | N/A | 88.89s | 108.61s |
| SWE-bench | $0.78 | $0.01 | 5m12s | 4m04s |
| Vibe Code Bench v1.1 | $3.83 | $0.07 | 20m12s | 26m20s |
Results available only for Gemini 3.1 Pro Preview (02/26)
- Vals Index
- BioMysteryBench
- MysteryMechanism
- Terminal-Bench Science
- IOI
- ProgramBench
- Terminal-Bench 4.0
- Vibe Code Bench 1-100
- Public Benefits Bench v1.1
Results available only for MiMo V2.5
None.