Ministral-3-3B-Base-2512: benchmark results
Mistral's December 2025 pretrained Ministral 3 3B checkpoint, the base its Instruct and Reasoning releases are tuned from. Provider: Mistral. Released 2025-12-02. Access: Open.
Aggregate Rating 1484 ± 24, rank #2604 of 10916 rated models, from 268 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Polish NLU - Polemo2 | 98.75 | Matthews correlation coefficient (%) | 100 |
| EuroEval Catalan NLU - MultiWikiQA CA | 78.35 | Character F1 (%) | 99.1 |
| EuroEval Estonian NLU - MultiWikiQA ET | 77.34 | Character F1 (%) | 98.8 |
| EuroEval Icelandic NLU - NQII | 62.37 | Character F1 (%) | 98.2 |
| EuroEval Catalan NLU - Guia CAT | 75.72 | Matthews correlation coefficient (%) | 98.1 |
| EuroEval German NLU - Germanquad | 68.77 | Character F1 (%) | 96.9 |
| EuroEval Finnish NLU - Tydiqa FI | 76.11 | Character F1 (%) | 96.2 |
| EuroEval Slovene NLU - Sentinews | 54.45 | Matthews correlation coefficient (%) | 96.1 |
| EuroEval Czech NLU - SQAD | 70.84 | Character F1 (%) | 95.6 |
| EuroEval German Summarization - Mlsum DE | 38.47 | chrF3++ (%) | 95.3 |
| EuroEval Catalan NLU - WikiANN CA | 74.81 | Micro F1 excluding MISC (%) | 94.3 |
| EuroEval Portuguese NLU - SST-2 PT | 85.31 | Matthews correlation coefficient (%) | 94.1 |
Interactive version: theaggregate.ai/model?slug=ministral-3-3b-base-2512 · How It Works · Data refreshed daily, snapshot 2026-11-08.