mistral-maths7B: benchmark results
Provider: Mistral. Access: Open.
Aggregate Rating 1346 ± 35, rank #7810 of 10916 rated models, from 6 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open LLM Leaderboard v1 - TruthfulQA MC2 | 57.3 | MC2 (%) (0-shot) | 69.7 |
| Open LLM Leaderboard v1 - GSM8K | 48.82 | Accuracy (%) (5-shot) | 61.9 |
| Open LLM Leaderboard v1 - MMLU | 54.54 | Accuracy (%) (5-shot) | 34.1 |
| Open LLM Leaderboard v1 - WinoGrande | 72.45 | Accuracy (%) (5-shot) | 29.1 |
| Open LLM Leaderboard v1 - ARC Challenge | 52.05 | Normalized accuracy (%) (25-shot) | 26.6 |
| Open LLM Leaderboard v1 - HellaSwag | 74.77 | Normalized accuracy (%) (10-shot) | 25.3 |
Interactive version: theaggregate.ai/model?slug=mistral-maths7b · How It Works · Data refreshed daily, snapshot 2026-11-08.