Mistral 7B Instruct (v0.3): benchmark results
Mistral 7B Instruct v0.3 checkpoint. Provider: Mistral. Released 2024-05-22. Access: Open.
Unified ELO 1418 ± 1, rank #1076 of 1392 rated models, from 169 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| LA Leaderboard - AQuAS | 73.27 | Accuracy (%) | 100 |
| LA Leaderboard - Spanish Law Exams | 40.34 | Accuracy (%) | 93.4 |
| Enkrypt AI - Bias Risk | 51.42 | Risk Score | 92 |
| HumanLikeness - Word-2 | 25 | Humanlike Score (%) | 89.5 |
| LA Leaderboard - PIQA Catalan | 67.25 | Accuracy (%) | 83.8 |
| LA Leaderboard | 55.98 | Average Score (%) | 77.9 |
| MERA - RWSD | 63.85 | Accuracy (%) | 77.5 |
| SpeechMap Compliance | 78.3 | % Requests Completed | 76 |
| Enkrypt AI - Safety Risk | 21.81 | Risk Score | 75.2 |
| ChineseSafe Benchmark | 70.41 | Accuracy (%) | 72.2 |
| MERA - ruCodeEval | 29.21 | pass@1 (%) | 72.2 |
| LA Leaderboard - XNLI Galician | 46.68 | Accuracy (%) | 71.3 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-instruct-v0-3 · How It Works · Data refreshed daily, snapshot 2026-09-05.