← All comparisons
GPT-5.5 Instant (May 2026) vs Magistral Small 1.2
OpenAI vs Mistral — side-by-side benchmark comparison
| GPT-5.5 Instant (May 2026) | Magistral Small 1.2 | |
|---|---|---|
| Intelligence Index | 41.8 | 18.2 |
| Coding Index | 45.1 | 14.8 |
| Math Index | — | 80.3 |
| Output speed (tok/s) | 0.0 | 111.1 |
| Blended price ($/1M) | $11.25 | $0.75 |
| Time to first token (s) | 0.00s | 0.38s |
| aime | — | — |
| aime 25 | — | 80.3% |
| artificial analysis coding index | 45.10 | 14.80 |
| artificial analysis intelligence index | 41.80 | 18.20 |
| artificial analysis math index | — | 80.30 |
| gpqa | 84.6% | 66.3% |
| hle | 20.3% | 6.1% |
| ifbench | 71.5% | 44.4% |
| lcr | 55.7% | 16.3% |
| livecodebench | — | 72.3% |
| math 500 | — | — |
| mmlu pro | — | 76.8% |
| scicode | 50.3% | 35.2% |
| tau2 | 49.4% | 27.8% |
| terminalbench hard | 42.4% | 4.5% |
Benchmark data from Artificial Analysis.