Mistral — benchmark results
Provider: Mistral. Access: Open.
Unified ELO 895 ± 125, rank #1726 of 1726 rated models, from 14 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MMedBench | 60.73 | Accuracy (%) | 75 |
| MMedBench - Rationale (BLEU-1) | 45.37 | Accuracy (%) | 66.7 |
| MedS-Bench - Text Summarization | 24.48 | Score | 63.6 |
| ClinicBench | 37.81 | Average Score | 54.5 |
| MedS-Bench - Concept Explanation | 13.53 | Score | 54.5 |
| MedS-Bench - Text Classification | 48.73 | Score | 54.5 |
| MedS-Bench - Diagnosis | 34.8 | Score | 45.5 |
| MedS-Bench - Information Extraction | 70.18 | Score | 27.3 |
| MedS-Bench - Clinical Outcome Prediction | 50.14 | Score | 22.7 |
| BIRD-SQL | 55.84 | Execution Accuracy (%) | 22.2 |
| MedS-Bench - Treatment Planning | 38.93 | Score | 18.2 |
| RULER | 68.4 | Avg Accuracy (%) | 15 |
Interactive version: aibenchmarks.dev/model?slug=mistral · How the rankings work · Data refreshed daily, snapshot 2026-07-08.