Ministral-3-14B-Base-2512: benchmark results
Mistral's December 2025 pretrained Ministral 3 14B checkpoint, the base its Instruct and Reasoning releases are tuned from. Provider: Mistral. Released 2025-12-02. Access: Open.
Aggregate Rating 1568 ± 24, rank #1425 of 10916 rated models, from 268 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval English Summarization - CNN Dailymail | 46.05 | chrF3++ (%) | 100 |
| EuroEval Finnish NLU - Tydiqa FI | 78.61 | Character F1 (%) | 100 |
| EuroEval Italian NLU - SQuAD IT | 78.66 | Character F1 (%) | 99.8 |
| EuroEval Dutch NLU - SQuAD NL | 86.09 | Character F1 (%) | 99.7 |
| EuroEval French NLU - Allocine | 97.8 | Matthews correlation coefficient (%) | 99.7 |
| EuroEval Icelandic NLU - NQII | 65.72 | Character F1 (%) | 99.7 |
| EuroEval Danish NLU - Multi Wiki QA DA | 82.94 | Character F1 (%) | 99.6 |
| EuroEval Estonian NLU - MultiWikiQA ET | 77.94 | Character F1 (%) | 99.6 |
| EuroEval Lithuanian NLU - MultiWikiQA LT | 76.12 | Character F1 (%) | 99.6 |
| EuroEval Polish NLU - Polemo2 | 98.2 | Matthews correlation coefficient (%) | 99.6 |
| EuroEval Faroese NLU - FoQA | 79.94 | Character F1 (%) | 99.5 |
| EuroEval French NLU | 77.38 | NLU Average Score (%) | 99.5 |
Interactive version: theaggregate.ai/model?slug=ministral-3-14b-base-2512 · How It Works · Data refreshed daily, snapshot 2026-11-08.