Ministral-8B-Instruct-2410: benchmark results
October 2024 Mistral Ministral 8B Instruct checkpoint. Provider: Mistral. Released 2024-10-16. Access: Open.
Aggregate Rating 1431 ± 9, rank #1094 of 1656 rated models, from 412 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| Open CoT - LSAT Reading Comprehension | 23.05 | CoT Gain (%) | 94.7 |
| Galileo Tool Tasks - xLAM Single Tool Multiple Call | 98 | Accuracy (%) | 91.7 |
| LatamBoard - Spanish OpenBookQA | 38.8 | Score (%) | 90.9 |
| LatamBoard - Spanish XNLI | 48.71 | Score (%) | 90.9 |
| LatamBoard - Spanish COPA | 84 | Score (%) | 87.9 |
| Open CoT - LSAT Logical Reasoning | 18.63 | CoT Gain (%) | 87.8 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 75.35 | Character F1 (%) | 86.6 |
| EuroEval Spanish NLU - MLQA ES | 65.87 | Character F1 (%) | 86.6 |
| DoRA (Defense Documents) - Completeness (Oracle Evidence) | 83.69 | RAGEval keypoint completeness (%) of grounded answers to the | 85.7 |
| Mobile-MMLU | 71.5 | Overall Accuracy (%) | 85.7 |
| Open CoT Leaderboard | 12.79 | Average CoT Gain (%) | 85.5 |
| SpeechMap Compliance | 85.8 | % Requests Completed | 84.9 |
Interactive version: theaggregate.ai/model?slug=ministral-8b-instruct-2410 · How It Works · Data refreshed daily, snapshot 2026-10-20.