← All comparisons
Claude 4.1 Opus (Non-reasoning) vs Pixtral Large
Anthropic vs Mistral — side-by-side benchmark comparison
| Claude 4.1 Opus (Non-reasoning) | Pixtral Large | |
|---|---|---|
| Intelligence Index | 36.0 | 14.0 |
| Coding Index | — | — |
| Math Index | — | 2.3 |
| Output speed (tok/s) | 44.7 | 54.9 |
| Blended price ($/1M) | $32.81 | $3.00 |
| Time to first token (s) | 1.63s | 0.50s |
| aime | — | 7.0% |
| aime 25 | — | 2.3% |
| artificial analysis coding index | — | — |
| artificial analysis intelligence index | 36.00 | 14.00 |
| artificial analysis math index | — | 2.30 |
| gpqa | — | 50.5% |
| hle | — | 3.6% |
| ifbench | — | 34.5% |
| lcr | — | 10.3% |
| livecodebench | — | 26.1% |
| math 500 | — | 71.4% |
| mmlu pro | — | 70.1% |
| scicode | — | 29.2% |
| tau2 | — | 36.5% |
| terminalbench hard | — | — |
Benchmark data from Artificial Analysis.