Claude Opus 4.7 vs Kimi K3: Benchmark Comparison
Claude Opus 4.7 has the higher score on 5 of 20 shared benchmarks; Kimi K3 leads on 15.
The largest observed score gap is 27.78 pts on Code Migration , where Claude Opus 4.7 leads.
Reported ±1 standard-error ranges overlap on 5 of 19 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 | Claude Opus 4.7 | Kimi K3 | Gap | Reported uncertainty |
|---|---|---|---|---|
| Vals RSI Index | 19.08% | 20.88% | 1.80 pts | Uncertainty comparison unavailable |
| Harvey's Legal Agent Benchmark | 6.67% ±2.00 | 12.92% ±2.68 | 6.25 pts | Reported ±1 SE ranges do not overlap |
| Legal Research Bench | 38.46% ±3.38 | 46.15% ±3.46 | 7.69 pts | Reported ±1 SE ranges do not overlap |
| LegalBench | 85.25% ±0.45 | 86.21% ±0.41 | 0.96 pts | Reported ±1 SE ranges do not overlap |
| EMB | 63.73% ±3.11 | 66.68% ±2.87 | 2.95 pts | Reported ±1 SE ranges overlap |
| Finance Agent (v2) | 51.51% ±0.49 | 53.11% ±0.37 | 1.60 pts | Reported ±1 SE ranges do not overlap |
| MortgageTax | 70.27% ±0.90 | 66.34% ±0.91 | 3.93 pts | Reported ±1 SE ranges do not overlap |
| Tax Agent Bench | 63.86% ±3.14 | 68.67% ±3.08 | 4.81 pts | Reported ±1 SE ranges overlap |
| TaxEval v2 | 75.27% ±0.84 | 75.72% ±0.84 | 0.45 pts | Reported ±1 SE ranges overlap |
| MedCode | 54.86% ±2.21 | 49.36% ±2.20 | 5.50 pts | Reported ±1 SE ranges do not overlap |
| MedScribe | 82.95% ±1.98 | 88.05% ±1.98 | 5.09 pts | Reported ±1 SE ranges do not overlap |
| GPQA Diamond | 90.15% ±1.54 | 92.93% ±1.31 | 2.78 pts | Reported ±1 SE ranges overlap |
| MMLU Pro | 89.87% ±0.30 | 87.97% ±0.32 | 1.90 pts | Reported ±1 SE ranges do not overlap |
| MMMU Pro | 85.55% ±0.84 | 88.15% ±0.78 | 2.60 pts | Reported ±1 SE ranges do not overlap |
| SAGE | 56.10% ±3.37 | 52.78% ±3.41 | 3.32 pts | Reported ±1 SE ranges overlap |
| Code Migration | 43.88% ±4.22 | 16.10% ±4.10 | 27.78 pts | Reported ±1 SE ranges do not overlap |
| LiveCodeBench | 85.07% ±1.01 | 87.19% ±0.97 | 2.11 pts | Reported ±1 SE ranges do not overlap |
| ProgramBench | 0.00% ±0.00 | 2.00% ±0.99 | 2.00 pts | Reported ±1 SE ranges do not overlap |
| SWE-bench | 82.00% ±1.72 | 93.40% ±1.11 | 11.40 pts | Reported ±1 SE ranges do not overlap |
| Vibe Code Bench v1.1 | 71.00% ±4.51 | 84.97% ±2.75 | 13.97 pts | Reported ±1 SE ranges do not overlap |
Performance by category
| Category | Claude Opus 4.7 average | Kimi K3 average |
|---|---|---|
| Legal | 43.46% | 48.43% |
| Finance | 64.93% | 66.10% |
| Healthcare | 68.91% | 68.70% |
| Academic | 88.52% | 89.68% |
| Education | 56.10% | 52.78% |
| Coding | 56.39% | 56.73% |
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 | Claude Opus 4.7 cost | Kimi K3 cost | Claude Opus 4.7 latency | Kimi K3 latency |
|---|---|---|---|---|
| Vals RSI Index | $1002.72 | $136.82 | 90h00m | 90h00m |
| Harvey's Legal Agent Benchmark | $10.81 | $3.83 | 24m44s | 13m58s |
| Legal Research Bench | $3.06 | $3.47 | 8m32s | 12m40s |
| LegalBench | N/A | N/A | 26.62s | 7.11s |
| EMB | $12.76 | $3.07 | 38m36s | 12m21s |
| Finance Agent (v2) | $4.03 | $1.91 | 6m00s | 4m46s |
| MortgageTax | N/A | N/A | 16.10s | 78.94s |
| Tax Agent Bench | $1.27 | $2.79 | 4m39s | 40m15s |
| TaxEval v2 | N/A | N/A | 62.07s | 2m25s |
| MedCode | N/A | N/A | 54.25s | 36.99s |
| MedScribe | N/A | N/A | 67.50s | 38.24s |
| GPQA Diamond | N/A | N/A | 3m23s | 2m02s |
| MMLU Pro | N/A | N/A | 3m08s | 38.72s |
| MMMU Pro | N/A | N/A | 70.80s | 106.52s |
| SAGE | N/A | N/A | 2m05s | 47.30s |
| Code Migration | $36.39 | $13.87 | 1h12m | 4h18m |
| LiveCodeBench | N/A | N/A | 7m42s | 3m20s |
| ProgramBench | N/A | N/A | 54m07s | 5h42m |
| SWE-bench | $2.42 | $0.76 | 7m22s | 10m19s |
| Vibe Code Bench v1.1 | $21.41 | $10.01 | 35m42s | 16m39s |
Results available only for Claude Opus 4.7
None.
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
- Public Benefits Bench v1.1
- Time Horizon Index: KSP