Gemini 3.1 Pro Preview (02/26) vs Kimi K2.6: Benchmark Comparison
Gemini 3.1 Pro Preview (02/26) has the higher score on 9 of 20 shared benchmarks; Kimi K2.6 leads on 10.
The largest observed score gap is 18.92 pts on MedCode , where Gemini 3.1 Pro Preview (02/26) leads.
Reported ±1 standard-error ranges overlap on 9 of 20 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) | Kimi K2.6 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 0.00% ±0.00 | 1.67% ±0.83 | 1.67 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 | 84.74% ±0.45 | 2.66 pts | Reported ±1 SE ranges do not overlap |
| EMB | 52.62% ±2.97 | 57.85% ±2.74 | 5.24 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 42.98% ±1.21 | 44.90% ±0.73 | 1.92 pts | Reported ±1 SE ranges overlap |
| MortgageTax | 69.40% ±0.91 | 65.82% ±0.93 | 3.58 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 37.84% ±3.10 | 45.13% ±3.23 | 7.29 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 72.88% ±0.86 | 74.65% ±0.85 | 1.77 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 59.06% ±2.00 | 40.14% ±2.04 | 18.92 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 76.11% ±1.92 | 78.15% ±1.79 | 2.03 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 95.45% ±1.05 | 89.14% ±2.00 | 6.31 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 90.99% ±0.28 | 87.57% ±0.33 | 3.41 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 88.21% ±0.78 | 86.30% ±0.83 | 1.91 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 48.68% ±3.29 | 50.22% ±3.43 | 1.55 pts | Reported ±1 SE ranges overlap |
| Code Migration | 17.31% ±3.93 | 27.77% ±4.14 | 10.46 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 88.48% ±0.93 | 86.77% ±0.97 | 1.71 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 0.00% ±0.00 | 0.00% ±0.00 | 0.00 pts | Reported ±1 SE ranges overlap |
| SWE-bench | 78.80% ±1.83 | 76.20% ±1.91 | 2.60 pts | Reported ±1 SE ranges overlap |
| Vibe Code Bench v1.1 | 32.03% ±4.34 | 37.89% ±4.91 | 5.86 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 53.79% ±1.30 | 56.63% ±1.29 | 2.84 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Gemini 3.1 Pro Preview (02/26) average | Kimi K2.6 average |
|---|---|---|
| Legal | 36.02% | 34.09% |
| Finance | 55.14% | 57.67% |
| Healthcare | 67.59% | 59.15% |
| Academic | 91.55% | 87.67% |
| Education | 48.68% | 50.22% |
| Coding | 43.33% | 45.73% |
| Social Mobility | 53.79% | 56.63% |
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 | Kimi K2.6 cost | Gemini 3.1 Pro Preview (02/26) latency | Kimi K2.6 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $1.15 | $0.82 | 7m17s | 23m23s |
| Legal Research Bench | $1.10 | $1.52 | 4m02s | 30m53s |
| LegalBench | N/A | N/A | 10.06s | 50.07s |
| EMB | $3.77 | $2.20 | 11m49s | 39m48s |
| Finance Agent (v2) | $1.59 | $1.14 | 3m28s | 27m10s |
| MortgageTax | N/A | N/A | 23.62s | 109.76s |
| Tax Agent Bench | $0.30 | $0.57 | 100.96s | 15m24s |
| TaxEval v2 | N/A | N/A | 41.93s | 4m27s |
| MedCode | N/A | N/A | 38.52s | 5m05s |
| MedScribe | N/A | N/A | 69.14s | 8m14s |
| GPQA Diamond | N/A | N/A | 65.76s | 9m05s |
| MMLU Pro | N/A | N/A | 23.91s | 2m36s |
| MMMU Pro | N/A | N/A | 76.99s | 4m02s |
| SAGE | N/A | N/A | 61.17s | 4m31s |
| Code Migration | $1.61 | $5.11 | 12m01s | 1h30m |
| LiveCodeBench | N/A | N/A | 88.89s | 8m49s |
| ProgramBench | N/A | N/A | 12m36s | 1h20m |
| SWE-bench | $0.78 | $0.27 | 5m12s | 8m40s |
| Vibe Code Bench v1.1 | $3.83 | $1.93 | 20m12s | 49m28s |
| Public Benefits Bench v1.1 | $0.38 | $0.91 | 3m06s | 9m55s |
Results available only for Gemini 3.1 Pro Preview (02/26)
- Vals Index
- ProofBench v1.1
- BioMysteryBench
- MysteryMechanism
- Terminal-Bench Science
- IOI
- Terminal-Bench 4.0
- Vibe Code Bench 1-100
Results available only for Kimi K2.6
None.