Mixtral 8x7B 32K: benchmark results
Provider: Mistral. Released 2023-12-11.
Aggregate Rating 1471 ± 24, rank #2818 of 10916 rated models, from 86 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| HELM Lite - LegalBench - International Citizenship Questions | 61.2 | Exact Match (%) | 85 |
| HELM Lite - LegalBench - Function of Decision Section | 42.78 | Exact Match (%) | 76.7 |
| HELM NarrativeQA | 76.66 | F1 (%) | 76.7 |
| HELM Lite - WMT 2014 (fr-en) | 23.04 | BLEU-4 (%) | 73 |
| HELM NaturalQuestions (Closed) | 42.73 | F1 (%) | 71.1 |
| HELM MMLU - Human Sexuality | 87.02 | EM | 67.9 |
| HELM MMLU - Us Foreign Policy | 93 | EM | 62.8 |
| HELM MMLU - World Religions | 87.13 | EM | 62.2 |
| HELM Lite | 56.26 | Mean win rate (imported) | 61.8 |
| HELM MMLU - College Computer Science | 67 | EM | 59 |
| HELM Lite - LegalBench - Corporate Lobbying | 77.35 | Exact Match (%) | 57.8 |
| HELM MMLU - Computer Security | 81 | EM | 57.1 |
Interactive version: theaggregate.ai/model?slug=mixtral-8x7b-32k · How It Works · Data refreshed daily, snapshot 2026-11-08.