Llama-Poro-2-8B-Instruct: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1486 ± 43, rank #2511 of 10552 rated models, from 23 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Finnish NLU - Scandisent FI | 93.36 | Matthews correlation coefficient (%) | 98.7 |
| EuroEval Finnish NLU - ScaLA FI | 29.3 | Matthews correlation coefficient (%) | 72 |
| EuroEval Estonian NLU - Estonian Valence | 52.84 | Matthews correlation coefficient (%) | 67.2 |
| EuroEval Finnish NLU | 57.95 | NLU Average Score (%) | 65.7 |
| EuroEval Finnish | 52.95 | Average Score (%) | 64.6 |
| EuroEval Portuguese NLU - ScaLA PT | 18.15 | Matthews correlation coefficient (%) | 63.9 |
| EuroEval Portuguese NLU - SST-2 PT | 79.14 | Matthews correlation coefficient (%) | 61.4 |
| EuroEval Portuguese NLU | 51.91 | NLU Average Score (%) | 61 |
| EuroEval Estonian NLU | 43.19 | NLU Average Score (%) | 60.2 |
| EuroEval Estonian NLU - EstNER | 46.58 | Micro F1 excluding MISC (%) | 59.4 |
| EuroEval Finnish Common Sense Reasoning | 32.96 | Common Sense Reasoning Average Score (%) | 56.7 |
| EuroEval Estonian NLU - Grammar ET | 13.11 | Matthews correlation coefficient (%) | 56.2 |
Interactive version: theaggregate.ai/model?slug=llama-poro-2-8b-instruct · How It Works · Data refreshed daily, snapshot 2026-11-16.