Devstral Small (May '25): benchmark results
Provider: Mistral. Released 2025-05-01. Access: Open.
Unified ELO 1490 ± 1, rank #911 of 1761 rated models, from 17 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA TAU-2 Bench | 38.01 | Accuracy (%) | 46 |
| AA Omniscience - Humanities & Social Sciences | 17.8 | Accuracy (%) | 41.3 |
| AA Omniscience - Law | 10 | Accuracy (%) | 40.6 |
| Artificial Analysis Intelligence Index | 6.04 | Intelligence Index | 37.3 |
| AA Omniscience - Health | 17.3 | Accuracy (%) | 37.2 |
| AA Terminal-Bench Hard | 6.06 | Accuracy (%) | 33.9 |
| AA Long Context Reasoning | 32 | Accuracy (%) | 33.6 |
| AA Omniscience - Software Engineering (SWE) | 20.2 | Accuracy (%) | 33.1 |
| AA-Omniscience Accuracy | 15.87 | Accuracy (%) | 29.4 |
| AA CritPt | 0 | Accuracy (%) | 24.4 |
| AA Omniscience | -56.78 | Score | 23.2 |
| AA Omniscience - Science, Engineering & Mathematics | 18.8 | Accuracy (%) | 20.6 |
Interactive version: theaggregate.ai/model?slug=devstral-small-may-25 · How It Works · Data refreshed daily, snapshot 2026-09-05.