Mistral-Small-24B-Base-2501: benchmark results
Mistral's 24B Apache-2.0 pretrained base behind Mistral Small 3, a knowledge-dense small model with a 32K context. Provider: Mistral. Released 2025-01-30. Access: Open.
Aggregate Rating 1532 ± 29, rank #670 of 1656 rated models, from 54 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard - GPQA | 38.76 | Score | 96.9 |
| Open Portuguese LLM - OAB Exams | 65.47 | Accuracy (%) | 96.8 |
| Open LLM Leaderboard - MMLU-Pro | 54.06 | Score | 96.6 |
| Open PL LLM - Belebele (multiple choice, 5-shot) | 91.78 | Accuracy (%) | 95.6 |
| Open PL LLM - Belebele (generative, 5-shot) | 91.78 | Accuracy (%) | 95 |
| Open Portuguese LLM - BLUEX | 73.99 | Accuracy (%) | 95 |
| Open Portuguese LLM - ENEM | 80.27 | Accuracy (%) | 93.4 |
| Open PL LLM - PoQuAD Open Book (generative, 5-shot) | 71.58 | Levenshtein Similarity (%) | 93.1 |
| Open LLM Leaderboard - BBH | 64.42 | Score | 89.5 |
| Open PL LLM - PolQA Open Book (generative, 5-shot) | 90.71 | Levenshtein Similarity (%) | 88.3 |
| Open PL LLM - PPC (multiple choice, 5-shot) | 79.1 | Accuracy (%) | 87.5 |
| Open PL LLM - 8TAGS (generative, 5-shot) | 78.18 | Accuracy (%) | 86.5 |
Interactive version: theaggregate.ai/model?slug=mistral-small-24b-base-2501 · How It Works · Data refreshed daily, snapshot 2026-10-20.