← All comparisons
Claude 4.1 Opus (Non-reasoning) vs Mistral Medium 3
Anthropic vs Mistral — side-by-side benchmark comparison
| Claude 4.1 Opus (Non-reasoning) | Mistral Medium 3 | |
|---|---|---|
| Intelligence Index | 36.0 | 18.8 |
| Coding Index | — | 13.6 |
| Math Index | — | 30.3 |
| Output speed (tok/s) | 44.7 | 36.0 |
| Blended price ($/1M) | $32.81 | $0.80 |
| Time to first token (s) | 1.63s | 0.61s |
| aime | — | 44.0% |
| aime 25 | — | 30.3% |
| artificial analysis coding index | — | 13.60 |
| artificial analysis intelligence index | 36.00 | 18.80 |
| artificial analysis math index | — | 30.30 |
| gpqa | — | 57.8% |
| hle | — | 4.3% |
| ifbench | — | 39.3% |
| lcr | — | 28.0% |
| livecodebench | — | 40.0% |
| math 500 | — | 90.7% |
| mmlu pro | — | 76.0% |
| scicode | — | 33.1% |
| tau2 | — | 24.3% |
| terminalbench hard | — | 3.8% |
Benchmark data from Artificial Analysis.