Llama 3 8B — benchmark results

Provider: Meta. Released 2024-04-18. Access: Open.

Unified ELO 1348 ± 15, rank #1421 of 1776 rated models, from 229 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
EuroEval Portuguese NLU - MultiWikiQA PT77.31Reading comprehension Score (%)93.8
EuroEval Finnish NLU - Tydiqa FI74.51Reading comprehension Score (%)92.5
EuroEval Dutch NLU - DBRD91.87Sentiment classification Score (%)90.2
MixRea18.3Implicit Information Ignorance Rate (I3R) (self-reported)90
EuroEval Finnish NLU - Scandisent FI92.2Sentiment classification Score (%)88.3
EuroEval French NLU - FQuAD72.66Reading comprehension Score (%)87.4
EuroEval Spanish NLU - MLQA ES65.7Reading comprehension Score (%)85.8
Open Medical LLM - MMLU Medical Genetics83Accuracy (%)85.8
LIBRA - LibrusecMHQA46.09Dataset Total Score (%)85.7
Open Medical LLM - MedMCQA57.47Accuracy (%)83.8
Open Medical LLM - MMLU Anatomy68.89Accuracy (%)83.5
IFEval Leaderboard78.84Final Score80

Interactive version: theaggregate.ai/model?slug=llama-3-8b · How the rankings work · Data refreshed daily, snapshot 2026-07-22.