Mixtral-8x22B-v0.1: benchmark results
Mistral's base (non-instruct) sparse mixture-of-experts with 141B total and 39B active parameters and a 64K context, under Apache 2.0 (April 2024). Provider: Mistral. Released 2024-04-10. Access: Open.
Aggregate Rating 1521 ± 22, rank #711 of 1656 rated models, from 46 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open PL LLM - PolQA Open Book (generative, 5-shot) | 92.21 | Levenshtein Similarity (%) | 98.1 |
| Open PL LLM - 8TAGS (generative, 5-shot) | 79.69 | Accuracy (%) | 96.2 |
| Open PL LLM - 8TAGS (multiple choice, 5-shot) | 79.76 | Accuracy (%) | 95.3 |
| Open PL LLM - Belebele (multiple choice, 5-shot) | 91.33 | Accuracy (%) | 94.1 |
| Open LLM Leaderboard - GPQA | 37.58 | Score | 94 |
| Open PL LLM - Belebele (generative, 5-shot) | 90.89 | Accuracy (%) | 91.5 |
| Open PL LLM - PSC (generative, 5-shot) | 97.94 | Binary F1 (%) | 90.9 |
| Open PL LLM - PPC (multiple choice, 5-shot) | 79.5 | Accuracy (%) | 90.6 |
| Open PL LLM - PoQuAD Open Book (generative, 5-shot) | 71.2 | Levenshtein Similarity (%) | 90.3 |
| RABBITS B4B | 78.82 | Accuracy (%) | 88.9 |
| Open PL LLM - PolQA Closed Book (generative, 5-shot) | 81.62 | Levenshtein Similarity (%) | 87.6 |
| Open PL LLM - PSC (multiple choice, 5-shot) | 86.69 | Binary F1 (%) | 86.2 |
Interactive version: theaggregate.ai/model?slug=mixtral-8x22b-v0-1 · How It Works · Data refreshed daily, snapshot 2026-10-20.