← All comparisons
Claude 4.5 Haiku (Reasoning) vs Kimi K2.6 (Non-reasoning)
Anthropic vs Kimi — side-by-side benchmark comparison
| Claude 4.5 Haiku (Reasoning) | Kimi K2.6 (Non-reasoning) | |
|---|---|---|
| Intelligence Index | 37.1 | 42.9 |
| Coding Index | 32.6 | 38.4 |
| Math Index | 83.7 | — |
| Output speed (tok/s) | 142.2 | 39.0 |
| Blended price ($/1M) | $2.19 | $1.71 |
| Time to first token (s) | 10.48s | 1.28s |
| aime | — | — |
| aime 25 | 83.7% | — |
| artificial analysis coding index | 32.60 | 38.40 |
| artificial analysis intelligence index | 37.10 | 42.90 |
| artificial analysis math index | 83.70 | — |
| gpqa | 67.2% | 78.8% |
| hle | 9.7% | 18.2% |
| ifbench | 54.3% | 44.3% |
| lcr | 70.3% | 57.7% |
| livecodebench | 61.5% | — |
| math 500 | — | — |
| mmlu pro | 76.0% | — |
| scicode | 43.3% | 39.5% |
| tau2 | 54.7% | 93.9% |
| terminalbench hard | 27.3% | 37.9% |
Benchmark data from Artificial Analysis.