Kimi K2.6 vs MiMo V2.5: Benchmark Comparison
Kimi K2.6 has the higher score on 16 of 18 shared benchmarks; MiMo V2.5 leads on 1.
The largest observed score gap is 15.97 pts on Tax Agent Bench , where Kimi K2.6 leads.
Reported ±1 standard-error ranges overlap on 3 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 | Kimi K2.6 | MiMo V2.5 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | 1.67% ±0.83 | 1.67% ±0.00 | 0.00 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 15.87% ±2.54 | 9.13% ±2.00 | 6.73 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 84.74% ±0.45 | 78.89% ±0.51 | 5.85 pts | Reported ±1 SE ranges do not overlap |
| EMB | 57.85% ±2.74 | 55.09% ±3.13 | 2.76 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 44.90% ±0.73 | 36.73% ±0.11 | 8.17 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 65.82% ±0.93 | 59.26% ±0.98 | 6.56 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 45.13% ±3.23 | 29.16% ±2.71 | 15.97 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 74.65% ±0.85 | 71.83% ±0.88 | 2.82 pts | Reported ±1 SE ranges do not overlap |
| MedCode | 40.14% ±2.04 | 31.89% ±2.02 | 8.25 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 78.15% ±1.79 | 72.15% ±1.85 | 6.00 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 89.14% ±2.00 | 81.57% ±2.05 | 7.58 pts | Reported ±1 SE ranges do not overlap |
| MMLU Pro | 87.57% ±0.33 | 82.93% ±0.37 | 4.64 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 86.30% ±0.83 | 80.00% ±0.96 | 6.30 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 50.22% ±3.43 | 43.27% ±3.39 | 6.96 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 27.77% ±4.14 | 14.23% ±3.69 | 13.54 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 86.77% ±0.97 | 81.51% ±1.07 | 5.26 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 76.20% ±1.91 | 71.00% ±2.03 | 5.20 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 37.89% ±4.91 | 42.17% ±4.57 | 4.28 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Kimi K2.6 average | MiMo V2.5 average |
|---|---|---|
| Legal | 34.09% | 29.90% |
| Finance | 57.67% | 50.41% |
| Healthcare | 59.15% | 52.02% |
| Academic | 87.67% | 81.50% |
| Education | 50.22% | 43.27% |
| Coding | 57.16% | 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 | Kimi K2.6 cost | MiMo V2.5 cost | Kimi K2.6 latency | MiMo V2.5 latency |
|---|---|---|---|---|
| Harvey's Legal Agent Benchmark | $0.82 | $0.04 | 23m23s | 6m59s |
| Legal Research Bench | $1.52 | $0.03 | 30m53s | 6m31s |
| LegalBench | N/A | N/A | 50.07s | 11.18s |
| EMB | $2.20 | $0.08 | 39m48s | 30m38s |
| Finance Agent (v2) | $1.14 | $0.09 | 27m10s | 5m39s |
| MortgageTax | N/A | N/A | 109.76s | 32.94s |
| Tax Agent Bench | $0.57 | $0.02 | 15m24s | 8m21s |
| TaxEval v2 | N/A | N/A | 4m27s | 26.30s |
| MedCode | N/A | N/A | 5m05s | 17.70s |
| MedScribe | N/A | N/A | 8m14s | 20.26s |
| GPQA Diamond | N/A | N/A | 9m05s | 89.39s |
| MMLU Pro | N/A | N/A | 2m36s | 30.25s |
| MMMU Pro | N/A | N/A | 4m02s | 42.47s |
| SAGE | N/A | N/A | 4m31s | 52.66s |
| Code Migration | $5.11 | $0.10 | 1h30m | 37m55s |
| LiveCodeBench | N/A | N/A | 8m49s | 108.61s |
| SWE-bench | $0.27 | $0.01 | 8m40s | 4m04s |
| Vibe Code Bench v1.1 | $1.93 | $0.07 | 49m28s | 26m20s |
Results available only for Kimi K2.6
- ProgramBench
- Public Benefits Bench v1.1
Results available only for MiMo V2.5
- ProofBench v1.1