Gemini 3.1 Pro Preview (02/26) vs MiMo V2.5 Pro: Benchmark Comparison
Gemini 3.1 Pro Preview (02/26) has the higher score on 11 of 18 shared benchmarks; MiMo V2.5 Pro leads on 7.
The largest observed score gap is 26.58 pts on MedCode , where Gemini 3.1 Pro Preview (02/26) leads.
Reported ±1 standard-error ranges overlap on 9 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 | Gemini 3.1 Pro Preview (02/26) | MiMo V2.5 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 33.44% ±1.13 | 33.84% ±1.17 | 0.40 pts | Reported ±1 SE ranges overlap |
| Harvey's Legal Agent Benchmark | 0.00% ±0.00 | 2.08% ±0.83 | 2.08 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 20.67% ±2.81 | 15.87% ±2.54 | 4.81 pts | Reported ±1 SE ranges overlap |
| LegalBench | 87.40% ±0.33 | 77.13% ±0.55 | 10.27 pts | Reported ±1 SE ranges do not overlap |
| EMB | 52.62% ±2.97 | 55.23% ±3.11 | 2.61 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 42.98% ±1.21 | 41.50% ±1.19 | 1.48 pts | Reported ±1 SE ranges overlap |
| Tax Agent Bench | 37.84% ±3.10 | 34.79% ±2.62 | 3.05 pts | Reported ±1 SE ranges overlap |
| TaxEval v2 | 72.88% ±0.86 | 73.79% ±0.86 | 0.91 pts | Reported ±1 SE ranges overlap |
| MedCode | 59.06% ±2.00 | 32.48% ±1.91 | 26.58 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 76.11% ±1.92 | 83.73% ±2.06 | 7.62 pts | Reported ±1 SE ranges do not overlap |
| ProofBench v1.1 | 26.00% ±4.41 | 22.00% ±4.16 | 4.00 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 95.45% ±1.05 | 82.58% ±2.47 | 12.88 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 90.99% ±0.28 | 84.59% ±0.45 | 6.39 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 17.31% ±3.93 | 21.56% ±4.13 | 4.24 pts | Reported ±1 SE ranges overlap |
| LiveCodeBench | 88.48% ±0.93 | 81.35% ±1.07 | 7.13 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 78.80% ±1.83 | 74.00% ±1.96 | 4.80 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 2.52% ±0.51 | 0.51% ±0.51 | 2.02 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 32.03% ±4.34 | 34.11% ±4.53 | 2.08 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Gemini 3.1 Pro Preview (02/26) average | MiMo V2.5 Pro average |
|---|---|---|
| Legal | 36.02% | 31.69% |
| Finance | 51.58% | 51.33% |
| Healthcare | 67.59% | 58.11% |
| Math | 26.00% | 22.00% |
| Academic | 93.22% | 83.59% |
| Coding | 43.83% | 42.31% |
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 Pro cost | Gemini 3.1 Pro Preview (02/26) latency | MiMo V2.5 Pro latency |
|---|---|---|---|---|
| Vals Index | $2.18 | $0.16 | 9m52s | 34m33s |
| Harvey's Legal Agent Benchmark | $1.15 | $0.11 | 7m17s | 9m04s |
| Legal Research Bench | $1.10 | $0.08 | 4m02s | 6m56s |
| LegalBench | N/A | N/A | 10.06s | 27.14s |
| EMB | $3.77 | $0.22 | 11m49s | 29m08s |
| Finance Agent (v2) | $1.59 | $0.21 | 3m28s | 8m17s |
| Tax Agent Bench | $0.30 | $0.04 | 100.96s | 6m17s |
| TaxEval v2 | N/A | N/A | 41.93s | 62.83s |
| MedCode | N/A | N/A | 38.52s | 38.45s |
| MedScribe | N/A | N/A | 69.14s | 90.71s |
| ProofBench v1.1 | $0.78 | $0.09 | 14m49s | 19m37s |
| GPQA Diamond | N/A | N/A | 65.76s | 4m11s |
| MMLU Pro | N/A | N/A | 23.91s | 67.76s |
| Code Migration | $1.61 | $0.24 | 12m01s | 48m53s |
| LiveCodeBench | N/A | N/A | 88.89s | 4m31s |
| SWE-bench | $0.78 | $0.02 | 5m12s | 7m51s |
| Terminal-Bench 4.0 | $4.11 | $0.23 | 18m44s | 2h07m |
| Vibe Code Bench v1.1 | $3.83 | $0.17 | 20m12s | 40m04s |
Results available only for Gemini 3.1 Pro Preview (02/26)
- MortgageTax
- BioMysteryBench
- MysteryMechanism
- Terminal-Bench Science
- MMMU Pro
- SAGE
- IOI
- ProgramBench
- Vibe Code Bench 1-100
- Public Benefits Bench v1.1
Results available only for MiMo V2.5 Pro
None.