Mistral's flagship proprietary model, tuned for European enterprise.
Mistral's flagship model competing with GPT-5 class models at a fraction of the cost
| Benchmark | Mistral Large 3 | Mistral Large | Δ |
|---|---|---|---|
| MMLU | 89.4% | 81.2% | +8.2 |
| HumanEval | 91.2% | 45% | +46.2 |
| MATH | 82.1% | — | — |
| MMMU | 76.8% | — | — |