Mistral
Devstral Small (Jul '25) is a Mistral AI model profile from a lineup that spans compact Ministral models, specialist coding models, reasoning-oriented Magistral releases, and larger frontier open models. The local data makes its multilingual, coding, cost, and latency tradeoffs easier to compare against other providers.
Introducing Mistral 3Rank #374 across 566
No rank
Rank #40 across 371
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
Streaming speed is not measured for this model yet.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 9.3 | #374 |
| Artificial Analysis Math Index | math | 29.3 | #187 |
| MMLU-Pro | reasoning | 62.2% | #271 |
| GPQA | reasoning | 41.4% | #443 |
| Humanity's Last Exam |
| reasoning |
| 3.7% |
| #500 |
| LiveCodeBench | coding | 25.4% | #251 |
| SciCode | coding, reasoning | 24.3% | #382 |
| MATH-500 | math | 63.5% | #157 |
| AIME | math | 0.3% | #180 |
| Blended Price | cost | $0.150/M | #40 |
| Input Price | cost | $0.100/M | #46 |
| Output Price | cost | $0.300/M | #44 |
| Value Index | cost, overall | 62.0 | #66 |