Ministral-3-8B-Base-2512: benchmark results
Mistral's December 2025 pretrained Ministral 3 8B checkpoint, the base its Instruct and Reasoning releases are tuned from. Provider: Mistral. Released 2025-12-02. Access: Open.
Aggregate Rating 1535 ± 24, rank #1891 of 10916 rated models, from 268 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Finnish NLU - Tydiqa FI | 77.99 | Character F1 (%) | 99.7 |
| EuroEval Estonian NLU - MultiWikiQA ET | 77.4 | Character F1 (%) | 99.2 |
| EuroEval Croatian NLU - MMS HR | 48.41 | Matthews correlation coefficient (%) | 99.1 |
| EuroEval Icelandic NLU - NQII | 63.52 | Character F1 (%) | 98.4 |
| EuroEval Polish NLU - Polemo2 | 97.11 | Matthews correlation coefficient (%) | 98.3 |
| EuroEval Italian NLU - SQuAD IT | 76.39 | Character F1 (%) | 98.1 |
| EuroEval German Summarization - Mlsum DE | 39.57 | chrF3++ (%) | 98 |
| EuroEval Dutch NLU - SQuAD NL | 84.01 | Character F1 (%) | 97.8 |
| EuroEval Czech NLU - CSFD Sentiment | 67.56 | Matthews correlation coefficient (%) | 97.4 |
| EuroEval Catalan NLU - WikiANN CA | 77.83 | Micro F1 excluding MISC (%) | 97.2 |
| EuroEval Greek NLU - MultiWikiQA EL | 73.21 | Character F1 (%) | 96.6 |
| EuroEval Greek Summarization - Greek Wikipedia | 32.63 | chrF3++ (%) | 96.6 |
Interactive version: theaggregate.ai/model?slug=ministral-3-8b-base-2512 · How It Works · Data refreshed daily, snapshot 2026-11-08.