Kimi K3 vs GPT 5.5: Benchmark Comparison
Kimi K3 has the higher score on 16 of 22 shared benchmarks; GPT 5.5 leads on 6.
The largest observed score gap is 29.06 pts on Code Migration , where GPT 5.5 leads.
Reported ±1 standard-error ranges overlap on 12 of 21 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 | GPT 5.5 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 20.88% | 16.48% | 4.40 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 12.92% ±2.68 | 3.75% ±1.17 | 9.17 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 46.15% ±3.46 | 40.38% ±3.41 | 5.77 pts | Reported ±1 SE ranges overlap |
| LegalBench | 86.21% ±0.41 | 86.52% ±0.41 | 0.30 pts | Reported ±1 SE ranges overlap |
| EMB | 66.68% ±2.87 | 64.54% ±2.87 | 2.13 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 53.11% ±0.37 | 51.76% ±0.55 | 1.35 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 66.34% ±0.91 | 68.76% ±0.91 | 2.42 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 68.67% ±3.08 | 60.46% ±3.22 | 8.22 pts | Reported ±1 SE ranges do not overlap |
| TaxEval v2 | 75.72% ±0.84 | 74.98% ±0.85 | 0.74 pts | Reported ±1 SE ranges overlap |
| MedCode | 49.36% ±2.20 | 49.10% ±2.19 | 0.26 pts | Reported ±1 SE ranges overlap |
| MedScribe | 88.05% ±1.98 | 86.87% ±1.93 | 1.18 pts | Reported ±1 SE ranges overlap |
| GPQA Diamond | 92.93% ±1.31 | 93.18% ±1.29 | 0.25 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 87.97% ±0.32 | 88.14% ±0.32 | 0.17 pts | Reported ±1 SE ranges overlap |
| MMMU Pro | 88.15% ±0.78 | 88.27% ±0.77 | 0.12 pts | Reported ±1 SE ranges overlap |
| SAGE | 52.78% ±3.41 | 51.53% ±3.95 | 1.25 pts | Reported ±1 SE ranges overlap |
| Code Migration | 16.10% ±4.10 | 45.16% ±4.16 | 29.06 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 87.19% ±0.97 | 85.30% ±1.02 | 1.89 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 |
| SWE-bench | 93.40% ±1.11 | 82.60% ±1.70 | 10.80 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 84.97% ±2.75 | 69.85% ±4.54 | 15.12 pts | Reported ±1 SE ranges do not overlap |
| Public Benefits Bench v1.1 | 68.20% ±1.21 | 60.89% ±1.27 | 7.31 pts | Reported ±1 SE ranges do not overlap |
| Time Horizon Index: KSP | 10.50% ±5.38 | 8.33% ±0.00 | 2.17 pts | Reported ±1 SE ranges overlap |
Performance by category
| Category | Kimi K3 average | GPT 5.5 average |
|---|---|---|
| Legal | 48.43% | 43.55% |
| Finance | 66.10% | 64.10% |
| Healthcare | 68.70% | 67.98% |
| Academic | 89.68% | 89.86% |
| Education | 52.78% | 51.53% |
| Coding | 56.73% | 56.68% |
| Social Mobility | 68.20% | 60.89% |
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 | GPT 5.5 cost | Kimi K3 latency | GPT 5.5 latency |
|---|---|---|---|---|
| Vals RSI Index | $136.82 | $857.16 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $3.83 | $4.60 | 13m58s | 12m14s |
| Legal Research Bench | $3.47 | $7.40 | 12m40s | 37m45s |
| LegalBench | N/A | N/A | 7.11s | 18.14s |
| EMB | $3.07 | $3.27 | 12m21s | 15m11s |
| Finance Agent (v2) | $1.91 | $4.15 | 4m46s | 11m02s |
| MortgageTax | N/A | N/A | 78.94s | 28.24s |
| Tax Agent Bench | $2.79 | $4.07 | 40m15s | 15m52s |
| TaxEval v2 | N/A | N/A | 2m25s | 95.83s |
| MedCode | N/A | N/A | 36.99s | 2m40s |
| MedScribe | N/A | N/A | 38.24s | 2m13s |
| GPQA Diamond | N/A | N/A | 2m02s | 112.33s |
| MMLU Pro | N/A | N/A | 38.72s | 42.11s |
| MMMU Pro | N/A | N/A | 106.52s | 54.15s |
| SAGE | N/A | N/A | 47.30s | 76.05s |
| Code Migration | $13.87 | $6.44 | 4h18m | 31m35s |
| LiveCodeBench | N/A | N/A | 3m20s | 2m47s |
| ProgramBench | $70.48 | $6.95 | 5h42m | 22m03s |
| SWE-bench | $0.76 | $1.36 | 10m19s | 7m06s |
| Vibe Code Bench v1.1 | $10.01 | $16.66 | 16m39s | 31m52s |
| Public Benefits Bench v1.1 | $0.91 | $3.97 | 7m19s | 39m37s |
| Time Horizon Index: KSP | $216.43 | $516.36 | N/A | N/A |
Results available only for Kimi K3
- Vals Index
- ProofBench v1.1
- BioMysteryBench
- Terminal-Bench Science
- IOI
- Terminal-Bench 4.0
- Vibe Code Bench 1-100
- CyberBench v1.1
Results available only for GPT 5.5
- SkillsBench
- SRE Bench