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

BenchmarkScoreMetricPercentile
EuroEval Finnish NLU - Tydiqa FI70.5Reading comprehension Score (%)84.7
EuroEval Portuguese NLU - MultiWikiQA PT73.66Reading comprehension Score (%)77.4
LA Leaderboard - GalCoLA52.08Accuracy (%)66.2
EuroEval Finnish NLU - Scandisent FI90.2Sentiment classification Score (%)63.8
EuroEval Dutch NLU - DBRD88.79Sentiment classification Score (%)60.7
Open Japanese LLM - Mbpp Pylint Check51.41Score (%)58.4
EuroEval Portuguese NLU - SST-2 PT78.39Sentiment classification Score (%)57.5
LA Leaderboard54.39Average Score (%)57.4
HELMET (128K)33.5Average Score54.8
MERA - ruHumanEval13.96pass@1 (%)53.8
Open Japanese LLM - CG17.07Score (%)53.4
MERA - ruModAr59.23EM (%)52.4

Interactive version: theaggregate.ai/model?slug=llama-3-2-3b · How It Works · Data refreshed daily, snapshot 2026-09-05.