Kimi K3 vs MiMo V2.6 Pro: Benchmark Comparison
Kimi K3 has the higher score on 9 of 18 shared benchmarks; MiMo V2.6 Pro leads on 9.
The largest observed score gap is 26.91 pts on Code Migration , where MiMo V2.6 Pro 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 | Kimi K3 | MiMo V2.6 Pro | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals Index | 50.30% ±0.99 | 55.20% ±1.18 | 4.90 pts | Reported ±1 SE ranges do not overlap |
| Harvey's Legal Agent Benchmark | 12.92% ±2.68 | 10.83% ±2.45 | 2.08 pts | Reported ±1 SE ranges overlap |
| Legal Research Bench | 46.15% ±3.46 | 47.12% ±3.47 | 0.96 pts | Reported ±1 SE ranges overlap |
| EMB | 66.68% ±2.87 | 62.86% ±3.14 | 3.82 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.11% ±0.37 | 57.34% ±0.57 | 4.23 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 68.67% ±3.08 | 64.94% ±3.21 | 3.73 pts | Reported ±1 SE ranges overlap |
| MedCode | 49.36% ±2.20 | 44.97% ±2.10 | 4.39 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 88.05% ±1.98 | 88.31% ±1.94 | 0.26 pts | Reported ±1 SE ranges overlap |
| ProofBench v1.1 | 87.00% ±3.38 | 70.00% ±4.61 | 17.00 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench Science | 1.43% ±1.43 | 2.86% ±2.01 | 1.43 pts | Reported ±1 SE ranges overlap |
| SAGE | 52.78% ±3.41 | 45.05% ±3.40 | 7.73 pts | Reported ±1 SE ranges do not overlap |
| Code Migration | 16.10% ±4.10 | 43.01% ±4.32 | 26.91 pts | Reported ±1 SE ranges do not overlap |
| IOI | 48.94% ±9.82 | 39.33% ±2.41 | 9.61 pts | Reported ±1 SE ranges overlap |
| ProgramBench | 2.00% ±0.99 | 0.50% ±0.50 | 1.50 pts | Reported ±1 SE ranges do not overlap |
| Terminal-Bench 4.0 | 17.17% ±0.51 | 31.31% ±3.07 | 14.14 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 84.97% ±2.75 | 85.22% ±3.39 | 0.25 pts | Reported ±1 SE ranges overlap |
| CyberBench v1.1 | 75.24% ±5.56 | 72.86% ±5.50 | 2.38 pts | Reported ±1 SE ranges overlap |
| Public Benefits Bench v1.1 | 68.20% ±1.21 | 68.94% ±1.20 | 0.74 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Kimi K3 average | MiMo V2.6 Pro average |
|---|---|---|
| Legal | 29.54% | 28.97% |
| Finance | 62.82% | 61.71% |
| Healthcare | 68.70% | 66.64% |
| Math | 87.00% | 70.00% |
| Science | 1.43% | 2.86% |
| Education | 52.78% | 45.05% |
| Coding | 33.84% | 39.88% |
| Cyber | 75.24% | 72.86% |
| Social Mobility | 68.20% | 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 | Kimi K3 cost | MiMo V2.6 Pro cost | Kimi K3 latency | MiMo V2.6 Pro latency |
|---|---|---|---|---|
| Vals Index | $6.38 | $0.41 | 1h08m | 1h07m |
| Harvey's Legal Agent Benchmark | $3.83 | $0.22 | 13m58s | 25m49s |
| Legal Research Bench | $3.47 | $0.18 | 12m40s | 30m16s |
| EMB | $3.07 | $0.33 | 12m21s | 55m14s |
| Finance Agent (v2) | $1.91 | $0.20 | 4m46s | 10m48s |
| Tax Agent Bench | $2.79 | $0.13 | 40m15s | 25m48s |
| MedCode | N/A | N/A | 36.99s | 2m57s |
| MedScribe | N/A | N/A | 38.24s | 2m53s |
| ProofBench v1.1 | $1.65 | $0.25 | 34m31s | 1h02m |
| Terminal-Bench Science | $21.30 | $0.62 | 4h49m | 3h12m |
| SAGE | N/A | N/A | 47.30s | 3m40s |
| Code Migration | $13.87 | $0.69 | 4h18m | 2h49m |
| IOI | $14.17 | $0.84 | 4h18m | 2h26m |
| ProgramBench | $70.48 | $0.73 | 5h42m | 3h00m |
| Terminal-Bench 4.0 | $12.02 | $0.50 | 3h09m | 2h42m |
| Vibe Code Bench v1.1 | $10.01 | $1.04 | 16m39s | 1h01m |
| CyberBench v1.1 | $2.13 | $0.09 | 30m44s | 27m41s |
| Public Benefits Bench v1.1 | $0.91 | $0.08 | 7m19s | 37m14s |
Results available only for Kimi K3
- Vals RSI Index
- LegalBench
- MortgageTax
- TaxEval v2
- BioMysteryBench
- GPQA Diamond
- MMLU Pro
- MMMU Pro
- LiveCodeBench
- SWE-bench
- Vibe Code Bench 1-100
- Time Horizon Index: KSP
Results available only for MiMo V2.6 Pro
- MysteryMechanism
- SRE Bench