xlm-roberta-large: benchmark results
Provider: Meta. Released 2019-11-05. Access: Open.
Aggregate Rating 1553 ± 34, rank #1665 of 10916 rated models, from 168 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Estonian NLU - EstNER | 71.53 | Micro F1 excluding MISC (%) | 100 |
| EuroEval Finnish NLU - Turku NER FI | 83.33 | Micro F1 excluding MISC (%) | 100 |
| EuroEval German NLU - GermEval | 80.14 | Micro F1 excluding MISC (%) | 100 |
| EuroEval Hungarian NLU - Szeged NER | 92.6 | Micro F1 excluding MISC (%) | 100 |
| EuroEval Italian NLU - MultiNERD IT | 91.13 | Micro F1 excluding MISC (%) | 100 |
| EuroEval Romanian NLU - RONEC | 78.55 | Micro F1 excluding MISC (%) | 100 |
| EuroEval Czech NLU - PONER | 71.38 | Micro F1 excluding MISC (%) | 99.6 |
| EuroEval Serbian NLU - UNER SR | 91.76 | Micro F1 excluding MISC (%) | 99.5 |
| EuroEval Portuguese NLU - HAREM | 74.89 | Micro F1 excluding MISC (%) | 99.3 |
| EuroEval Latvian NLU - Fullstack NER LV | 79.69 | Micro F1 excluding MISC (%) | 99.1 |
| EuroEval Lithuanian NLU - WikiANN LT | 77.9 | Micro F1 excluding MISC (%) | 99.1 |
| EuroEval Slovene NLU - SSJ500K NER | 80.11 | Micro F1 excluding MISC (%) | 99 |
Interactive version: theaggregate.ai/model?slug=xlm-roberta-large · How It Works · Data refreshed daily, snapshot 2026-11-08.