← All comparisons
Claude 4.1 Opus (Reasoning) vs Mistral Small 3.2
Anthropic vs Mistral — side-by-side benchmark comparison
| Claude 4.1 Opus (Reasoning) | Mistral Small 3.2 | |
|---|---|---|
| Intelligence Index | 42.0 | 15.1 |
| Coding Index | 36.5 | 13.3 |
| Math Index | 80.3 | 27.0 |
| Output speed (tok/s) | 44.5 | 133.0 |
| Blended price ($/1M) | $32.81 | $0.13 |
| Time to first token (s) | 8.55s | 0.36s |
| aime | — | 32.3% |
| aime 25 | 80.3% | 27.0% |
| artificial analysis coding index | 36.50 | 13.30 |
| artificial analysis intelligence index | 42.00 | 15.10 |
| artificial analysis math index | 80.30 | 27.00 |
| gpqa | 80.9% | 50.5% |
| hle | 11.9% | 4.3% |
| ifbench | 55.4% | 33.5% |
| lcr | 66.3% | 17.3% |
| livecodebench | 65.4% | 27.5% |
| math 500 | — | 88.3% |
| mmlu pro | 88.0% | 68.1% |
| scicode | 40.9% | 26.4% |
| tau2 | 71.4% | 29.5% |
| terminalbench hard | 34.3% | 6.8% |
Benchmark data from Artificial Analysis.