Mistral Small 4 (Non-reasoning): benchmark results
Provider: Mistral. Released 2026-03-16. Access: Open.
Aggregate Rating 1508 ± 20, rank #1068 of 2201 rated models, from 34 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) - Java | 22.22 | Accuracy (%) | 80.2 |
| AA Omniscience - Software Engineering (SWE) - R | 14.29 | Accuracy (%) | 70 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 20 | Accuracy (%) | 63 |
| AA Omniscience - Software Engineering (SWE) - Go | 18.75 | Accuracy (%) | 60.1 |
| AA Omniscience - Software Engineering (SWE) - Python | 20.1 | Accuracy (%) | 58.9 |
| AA Omniscience - Software Engineering (SWE) - Dart | 18 | Accuracy (%) | 51.5 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 17.78 | Accuracy (%) | 51.4 |
| AA-Omniscience Hallucination Rate | 78.01 | Hallucination Rate (%) | 51.1 |
| AA Omniscience - Software Engineering (SWE) - C | 30 | Accuracy (%) | 47.6 |
| AA Omniscience - Software Engineering (SWE) - Julia | 8 | Accuracy (%) | 46.1 |
| AA Omniscience - Software Engineering (SWE) - Swift | 29.17 | Accuracy (%) | 45.7 |
| AA CritPt | 0.29 | Accuracy (%) | 45.5 |
Interactive version: theaggregate.ai/model?slug=mistral-small-4-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-12.