Mistral-7B-v0.1: benchmark results
Mistral AI's original 7B base model, a transformer with grouped-query and sliding-window attention, released under Apache 2.0 in September 2023. Provider: Mistral. Released 2023-09-27. Access: Open.
Aggregate Rating 1372 ± 8, rank #1326 of 1656 rated models, from 394 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA Original Paper | 40 | Total Score (%; mean of 17 problem-solving and exam tasks, M | 100 |
| MERA Original Paper - ruMMLU | 67.6 | Accuracy (%; 5-shot, log-likelihood) | 100 |
| MERA Original Paper - ruModAr | 51.6 | Exact Match (%; greedy generation) | 100 |
| MERA Original Paper - ruMultiAr | 19.5 | Exact Match (%; 5-shot, greedy generation) | 100 |
| MERA Original Paper - ruOpenBookQA | 73.5 | Accuracy (%; 5-shot, log-likelihood) | 100 |
| MERA Original Paper - ruWorldTree | 81 | Accuracy (%; 5-shot, log-likelihood) | 100 |
| MMLU-by-task - Econometrics | 50 | Accuracy (%) | 99.8 |
| Enkrypt AI - Bias Risk | 13.44 | Risk Score | 99.3 |
| MMLU-by-task - College Chemistry | 52 | Accuracy (%) | 99.3 |
| MMLU-by-task - Medical Genetics | 74 | Accuracy (%) | 98.9 |
| MMLU-by-task - Virology | 55.42 | Accuracy (%) | 98.8 |
| HELM Classic - IMDB | 96.2 | Exact Match (%) | 98.5 |
Interactive version: theaggregate.ai/model?slug=mistral-7b-v0-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.