Mistral-7B-v0.3: benchmark results
Provider: Mistral. Released 2024-05-22. Access: Open.
Unified ELO 1394 ± 1, rank #1160 of 1392 rated models, from 110 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Finnish NLU - Tydiqa FI | 69.63 | Reading comprehension Score (%) | 83.6 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 74.69 | Reading comprehension Score (%) | 83.4 |
| EuroEval Spanish NLU - MLQA ES | 64.36 | Reading comprehension Score (%) | 79.1 |
| EuroEval Dutch NLU - DBRD | 90.5 | Sentiment classification Score (%) | 76.2 |
| LA Leaderboard - PIQA Catalan | 65.13 | Accuracy (%) | 75.7 |
| EuroEval Portuguese NLU - SST-2 PT | 80.72 | Sentiment classification Score (%) | 70.7 |
| MERA - ruHumanEval | 19.15 | pass@1 (%) | 63.9 |
| MERA - ruCodeEval | 19.39 | pass@1 (%) | 63.2 |
| EuroEval Italian NLU - SQuAD IT | 69.12 | Reading comprehension Score (%) | 63 |
| LA Leaderboard - EusExams Basque | 31.68 | Accuracy (%) | 59.6 |
| Open Italian LLM - MMLU-Pro (IT) | 24.9 | Accuracy (%) | 57.1 |
| EuroEval Finnish NLU - Scandisent FI | 89.52 | Sentiment classification Score (%) | 54.5 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-v0-3 · How It Works · Data refreshed daily, snapshot 2026-09-05.