RoLlama3.1-8B-Instruct-DPO: benchmark results
Provider: Other. Access: Open.
Aggregate Rating 1463 ± 34, rank #3155 of 10916 rated models, from 48 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Romanian NLU - RoSent | 95.51 | Matthews correlation coefficient (%) | 74 |
| EuroEval Portuguese NLU - ScaLA PT | 21.54 | Matthews correlation coefficient (%) | 73 |
| EuroEval Catalan NLU - Guia CAT | 61.26 | Matthews correlation coefficient (%) | 64.9 |
| EuroEval Italian NLU - Sentipolc16 | 56.24 | Matthews correlation coefficient (%) | 64.4 |
| EuroEval Romanian Common Sense Reasoning | 29.36 | Common Sense Reasoning Average Score (%) | 63.5 |
| EuroEval Catalan Common Sense Reasoning | 23.59 | Common Sense Reasoning Average Score (%) | 62.2 |
| EuroEval Portuguese NLU - HAREM | 44.59 | Micro F1 excluding MISC (%) | 60.6 |
| EuroEval Catalan NLU - ScaLA CA | 13.85 | Matthews correlation coefficient (%) | 60.2 |
| EuroEval Portuguese | 49.52 | Average Score (%) | 60.1 |
| EuroEval Spanish Common Sense Reasoning | 49.04 | Common Sense Reasoning Average Score (%) | 59.7 |
| EuroEval Italian Common Sense Reasoning | 45.96 | Common Sense Reasoning Average Score (%) | 59.5 |
| EuroEval Italian NLU - ScaLA IT | 23.22 | Matthews correlation coefficient (%) | 58.4 |
Interactive version: theaggregate.ai/model?slug=rollama3-1-8b-instruct-dpo · How It Works · Data refreshed daily, snapshot 2026-11-08.