EuroLLM-22B-2512: benchmark results
Provider: EuroLLM. Released 2025-12-01. Access: Open.
Aggregate Rating 1486 ± 25, rank #2585 of 10916 rated models, from 283 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Dutch Simplification - Duidelijke Taal | 59.48 | Score (%) | 100 |
| EuroEval Serbian NLU - MultiWikiQA SR | 67 | Character F1 (%) | 97.5 |
| EuroEval Finnish NLU - Tydiqa FI | 75.86 | Character F1 (%) | 94.9 |
| Open PL LLM - PPC (multiple choice, 5-shot) | 80.7 | Accuracy (%) | 94 |
| EuroEval German Summarization - Mlsum DE | 38 | chrF3++ (%) | 93.9 |
| EuroEval German NLU - Germanquad | 67.96 | Character F1 (%) | 93.8 |
| EuroEval Czech NLU - CSFD Sentiment | 63.94 | Matthews correlation coefficient (%) | 93.4 |
| EuroEval Latvian NLU - MultiWikiQA LV | 68.67 | Character F1 (%) | 93.1 |
| Open PL LLM - PSC (multiple choice, 5-shot) | 91.49 | Binary F1 (%) | 92.8 |
| EuroEval Lithuanian NLU - MultiWikiQA LT | 70.17 | Character F1 (%) | 91.7 |
| EuroEval Slovene NLU - MultiWikiQA SL | 69.47 | Character F1 (%) | 91.6 |
| EuroEval Slovak NLU - Multi Wiki QA SK | 70.59 | Character F1 (%) | 91.1 |
Interactive version: theaggregate.ai/model?slug=eurollm-22b-2512 · How It Works · Data refreshed daily, snapshot 2026-11-08.