← All comparisons
Claude 3.5 Sonnet (Oct '24) vs Kimi K2
Anthropic vs Kimi — side-by-side benchmark comparison
| Claude 3.5 Sonnet (Oct '24) | Kimi K2 | |
|---|---|---|
| Intelligence Index | 15.9 | 26.3 |
| Coding Index | 30.2 | 22.1 |
| Math Index | — | 57.0 |
| Output speed (tok/s) | 0.0 | 23.9 |
| Blended price ($/1M) | $6.56 | $1.04 |
| Time to first token (s) | 0.00s | 1.59s |
| aime | 15.7% | 69.3% |
| aime 25 | — | 57.0% |
| artificial analysis coding index | 30.20 | 22.10 |
| artificial analysis intelligence index | 15.90 | 26.30 |
| artificial analysis math index | — | 57.00 |
| gpqa | 59.9% | 76.6% |
| hle | 3.9% | 7.0% |
| ifbench | — | 41.5% |
| lcr | — | 51.0% |
| livecodebench | 38.1% | 55.6% |
| math 500 | 77.1% | 97.1% |
| mmlu pro | 77.2% | 82.4% |
| scicode | 36.6% | 34.5% |
| tau2 | — | 61.1% |
| terminalbench hard | — | 15.9% |
Benchmark data from Artificial Analysis.