RoMistral-7B-Instruct-DPO: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1408 ± 35, rank #5702 of 10916 rated models, from 48 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Catalan NLU - MultiWikiQA CA | 72.59 | Character F1 (%) | 82 |
| EuroEval Romanian NLU - RoSent | 96.04 | Matthews correlation coefficient (%) | 81.6 |
| EuroEval Romanian NLU - MultiWikiQA RO | 65.97 | Character F1 (%) | 68 |
| EuroEval Catalan NLU - Guia CAT | 61.82 | Matthews correlation coefficient (%) | 65.9 |
| EuroEval Spanish NLU - Sentiment Headlines ES | 42.98 | Matthews correlation coefficient (%) | 61.7 |
| EuroEval Portuguese NLU - SST-2 PT | 78.73 | Matthews correlation coefficient (%) | 59.5 |
| EuroEval Catalan NLU | 50.4 | NLU Average Score (%) | 53.1 |
| EuroEval Catalan NLU - ScaLA CA | 5.73 | Matthews correlation coefficient (%) | 51.2 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 68.9 | Character F1 (%) | 51.2 |
| EuroEval Portuguese NLU - HAREM | 41.61 | Micro F1 excluding MISC (%) | 50.5 |
| EuroEval Italian NLU - Sentipolc16 | 49.58 | Matthews correlation coefficient (%) | 49.7 |
| EuroEval Spanish NLU | 41.98 | NLU Average Score (%) | 49 |
Interactive version: theaggregate.ai/model?slug=romistral-7b-instruct-dpo · How It Works · Data refreshed daily, snapshot 2026-11-08.