Llama 3.2 3B: benchmark results
Provider: Meta. Released 2024-09-25. Access: Open.
Unified ELO 1378 ± 1, rank #1199 of 1392 rated models, from 144 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| EuroEval Finnish NLU - Tydiqa FI | 70.5 | Reading comprehension Score (%) | 84.7 |
| EuroEval Portuguese NLU - MultiWikiQA PT | 73.66 | Reading comprehension Score (%) | 77.4 |
| LA Leaderboard - GalCoLA | 52.08 | Accuracy (%) | 66.2 |
| EuroEval Finnish NLU - Scandisent FI | 90.2 | Sentiment classification Score (%) | 63.8 |
| EuroEval Dutch NLU - DBRD | 88.79 | Sentiment classification Score (%) | 60.7 |
| Open Japanese LLM - Mbpp Pylint Check | 51.41 | Score (%) | 58.4 |
| EuroEval Portuguese NLU - SST-2 PT | 78.39 | Sentiment classification Score (%) | 57.5 |
| LA Leaderboard | 54.39 | Average Score (%) | 57.4 |
| HELMET (128K) | 33.5 | Average Score | 54.8 |
| MERA - ruHumanEval | 13.96 | pass@1 (%) | 53.8 |
| Open Japanese LLM - CG | 17.07 | Score (%) | 53.4 |
| MERA - ruModAr | 59.23 | EM (%) | 52.4 |
Interactive version: theaggregate.ai/model?slug=llama-3-2-3b · How It Works · Data refreshed daily, snapshot 2026-09-05.