Mixtral 8x22B Instruct (v0.1): benchmark results
Mistral Mixtral 8x22B Instruct v0.1 checkpoint. Provider: Mistral. Released 2024-04-17. Access: Open.
Unified ELO 1520 ± 1, rank #598 of 1392 rated models, from 93 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| MERA - RCB | 61.19 | Accuracy (%) | 98.3 |
| Open PL LLM - Generative | 68.64 | Average Generative Score (%) | 94.8 |
| MERA - ruHHH | 89.33 | Accuracy (%) | 93.3 |
| Open LLM Leaderboard - GPQA | 16.44 | Score | 93.2 |
| Open PL LLM - RAG | 73.55 | Average RAG Score (%) | 90.9 |
| Open PL LLM Leaderboard | 65.23 | Average Score (%) | 90.9 |
| MERA Code - RealCodeJava | 36.58 | pass@1 (%) | 90.8 |
| BiGGen-Bench | 3.94 | Average Score (1-5) | 90.2 |
| Open PL LLM - Multiple Choice | 60.79 | Average Multiple-Choice Score (%) | 88.8 |
| MERA - BPS | 99.5 | Accuracy (%) | 86.1 |
| MERA - CheGeKa | 42.03 | F1 (%) | 85.6 |
| Open LLM Leaderboard - IFEval | 71.84 | Score | 85.3 |
Interactive version: theaggregate.ai/model?slug=mixtral-8x22b-instruct-v0-1 · How It Works · Data refreshed daily, snapshot 2026-09-05.