← All comparisons
GPT-4o (Aug '24) vs Mistral Small 3.2
OpenAI vs Mistral — side-by-side benchmark comparison
| GPT-4o (Aug '24) | Mistral Small 3.2 | |
|---|---|---|
| Intelligence Index | 18.6 | 15.1 |
| Coding Index | 16.6 | 13.3 |
| Math Index | — | 27.0 |
| Output speed (tok/s) | 117.5 | 133.0 |
| Blended price ($/1M) | $4.38 | $0.13 |
| Time to first token (s) | 0.60s | 0.36s |
| aime | 11.7% | 32.3% |
| aime 25 | — | 27.0% |
| artificial analysis coding index | 16.60 | 13.30 |
| artificial analysis intelligence index | 18.60 | 15.10 |
| artificial analysis math index | — | 27.00 |
| gpqa | 52.1% | 50.5% |
| hle | 2.9% | 4.3% |
| ifbench | 36.0% | 33.5% |
| lcr | 35.0% | 17.3% |
| livecodebench | 31.7% | 27.5% |
| math 500 | 79.5% | 88.3% |
| mmlu pro | — | 68.1% |
| scicode | 33.1% | 26.4% |
| tau2 | 28.9% | 29.5% |
| terminalbench hard | 8.3% | 6.8% |
Benchmark data from Artificial Analysis.