Mixtral 8x7B Instruct (v0.1) — benchmark results
Mistral's Apache-2.0 sparse mixture-of-experts instruct model (December 2023) with 46.7B parameters, 12.9B active per token. Provider: Mistral. Released 2023-12-11. Access: Open.
Unified ELO 1423 ± 10, rank #1113 of 1776 rated models, from 135 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| European LLM Leaderboard - Zero-Shot Accuracy | 58.29 | Average Accuracy (%) | 100 |
| European LLM Leaderboard - Accuracy | 59.11 | Average Accuracy (%) | 94.1 |
| European LLM Leaderboard - FLORES200 | 17.31 | Average FLORES200 Score | 93.8 |
| European LLM Leaderboard - FLORES200 Target | 17.31 | FLORES200 Target Score | 93.8 |
| European LLM Leaderboard - FLORES200 Source | 17.31 | FLORES200 Source Score | 93.3 |
| Open CoT - LSAT Analytical Reasoning | 8.7 | CoT Gain (%) | 92.4 |
| AlpacaEval 1.0 | 94.78 | Win Rate (%) | 91.6 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 49.5 | Sentiment classification Score (%) | 89.6 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 75.38 | Reading comprehension Score (%) | 86.9 |
| Open CoT - LogiQA | 6.55 | CoT Gain (%) | 82.1 |
| EuroEval Spanish NLU - ScaLA ES | 33.06 | Linguistic acceptability Score (%) | 81.9 |
| EuroEval Danish Knowledge | 82.1 | Knowledge Average Score (%) | 80.1 |
Interactive version: theaggregate.ai/model?slug=mixtral-8x7b-instruct-v0-1 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.