Llama 3 8B Flashback V1: benchmark results
Provider: Meta. Access: Open.
Aggregate Rating 1380 ± 39, rank #6859 of 10916 rated models, from 16 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Portuguese NLU - MultiWikiQA PT | 72.38 | Character F1 (%) | 69.6 |
| EuroEval Portuguese NLU - SST-2 PT | 79.34 | Matthews correlation coefficient (%) | 61.8 |
| EuroEval Portuguese NLU - HAREM | 43.95 | Micro F1 excluding MISC (%) | 59.4 |
| EuroEval Danish Knowledge | 57.37 | Knowledge Average Score (%) | 53.5 |
| EuroEval Danish Knowledge - Danish Citizen Tests | 57.37 | MCC (x100) | 51.4 |
| EuroEval Norwegian Knowledge - NRK Quiz QA | 26.3 | MCC (x100) | 48.3 |
| EuroEval Norwegian Knowledge - Idioms NO | 3.67 | MCC (x100) | 48.2 |
| EuroEval Norwegian Knowledge | 14.98 | Knowledge Average Score (%) | 47.8 |
| EuroEval Portuguese NLU | 49.01 | NLU Average Score (%) | 47.6 |
| EuroEval English Knowledge | 70.6 | Knowledge Average Score (%) | 45.5 |
| EuroEval Norwegian Common Sense Reasoning | 40.52 | Common Sense Reasoning Average Score (%) | 43.2 |
| EuroEval Portuguese | 39.5 | Average Score (%) | 42.2 |
Interactive version: theaggregate.ai/model?slug=llama-3-8b-flashback-v1 · How It Works · Data refreshed daily, snapshot 2026-11-08.