Codestral (2501): benchmark results
Provider: Mistral. Released 2025-01-13.
Aggregate Rating 1555 ± 23, rank #1636 of 10916 rated models, from 12 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| DuckDB-NSQL | 70.7 | Execution Accuracy (%) | 88 |
| ResearchCodeBench | 33.49 | Task Success Rate (%) | 41.9 |
| Wolfram LLM Benchmarking Project | 34.3 | Correct Functionality (%) | 34.1 |
| OpenAI HumanEval | 82.32 | Accuracy (%) | 31.8 |
| AutoBaxBench - Easy - Correct & Secure | 9.52 | Correct and secure solutions, sec_pass@1 (%) | 30.8 |
| AutoBaxBench - Hard - Correct | 11.19 | Correct solutions, pass@1 (%) | 30.8 |
| AutoBaxBench - Medium - Correct | 21.43 | Correct solutions, pass@1 (%) | 30.8 |
| AutoBaxBench - Medium - Correct & Secure | 10.12 | Correct and secure solutions, sec_pass@1 (%) | 30.8 |
| AutoBaxBench - Easy - Correct | 35.95 | Correct solutions, pass@1 (%) | 23.1 |
| AutoBaxBench - Hard - Correct & Secure | 0.48 | Correct and secure solutions, sec_pass@1 (%) | 15.4 |
| RepairBench | 18.7 | Plausible@1 (%) | 7.9 |
| Defects4J | 19.8 | Defects4J Plausible @1 (imported) | 5.9 |
Interactive version: theaggregate.ai/model?slug=codestral-2501 · How It Works · Data refreshed daily, snapshot 2026-11-08.