Codestral-22B-v0.1: benchmark results
Mistral's 22B code-generation model (May 2024) covering 80+ programming languages with fill-in-the-middle, under the non-production MNPL license. Provider: Mistral. Released 2024-05-29. Access: Open.
Aggregate Rating 1465 ± 19, rank #911 of 1656 rated models, from 37 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open Portuguese LLM - FaQuAD NLI | 83.28 | Macro F1 (%) | 96.6 |
| McEval | 50.5 | AVG (%) | 81 |
| RoundTripCodeEval - Run-Length Encoding | 30.68 | Mean of exact match, edit similarity and pass@5 (%) over the | 78.6 |
| EvalPlus (HumanEval+ & MBPP+) | 67.8 | Pass@1 avg (%) | 75.8 |
| RoundTripCodeEval - LZW | 7.77 | Mean of exact match, edit similarity and pass@5 (%) over the | 71.4 |
| Open LLM Leaderboard - IFEval | 57.72 | Score | 69.7 |
| XRepoTest - PHP - Line Coverage | 23.01 | Focal line coverage (%; of all focal functions) | 69.2 |
| Open Portuguese LLM - ASSIN2 RTE | 91.26 | Macro F1 (%) | 65.7 |
| BigCodeBench | 41.8 | Pass@1 (%) | 64.8 |
| Open Portuguese LLM - OAB Exams | 45.83 | Accuracy (%) | 62.4 |
| Open Portuguese LLM - BLUEX | 54.8 | Accuracy (%) | 61.4 |
| Open LLM Leaderboard - MuSR | 41.87 | Score | 59.2 |
Interactive version: theaggregate.ai/model?slug=codestral-22b-v0-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.