ModernBERT-base: benchmark results
Provider: Answer.AI & LightOn. Released 2024-12-19. Access: Open.
Aggregate Rating 1308 ± 33, rank #8591 of 10916 rated models, from 180 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Hungarian NLU - Szeged NER | 69.12 | Micro F1 excluding MISC (%) | 78.7 |
| EuroEval Ukrainian NLU - NER UK | 56.05 | Micro F1 excluding MISC (%) | 67.5 |
| EuroEval Faroese NLU - FONE | 67.19 | Micro F1 excluding MISC (%) | 63.5 |
| EuroEval English NLU - CoNLL EN | 71.65 | Micro F1 excluding MISC (%) | 57.5 |
| EuroEval Bulgarian NLU - BG NER Bsnlp | 64.75 | Micro F1 excluding MISC (%) | 50.7 |
| EuroEval Spanish NLU - CoNLL ES | 60.66 | Micro F1 excluding MISC (%) | 50.7 |
| EuroEval Serbian NLU - UNER SR | 61.4 | Micro F1 excluding MISC (%) | 47.5 |
| EuroEval Italian NLU - MultiNERD IT | 59.27 | Micro F1 excluding MISC (%) | 47.3 |
| EuroEval Albanian NLU - ScaLA SQ | 5.03 | Matthews correlation coefficient (%) | 44 |
| EuroEval Greek NLU - ELNER | 49.8 | Micro F1 excluding MISC (%) | 43.8 |
| EuroEval Icelandic NLU - Hotter and Colder Sentiment | 26.72 | Matthews correlation coefficient (%) | 43.6 |
| EuroEval Catalan NLU - ScaLA CA | 1.21 | Matthews correlation coefficient (%) | 41.7 |
Interactive version: theaggregate.ai/model?slug=modernbert-base · How It Works · Data refreshed daily, snapshot 2026-11-08.