Meta Llama 3 8B Instruct — Benchmarks

Benchmark scores for Meta Llama 3 8B Instruct aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #103 of 123 open modelscomposite 28.3/100 across 9 benchmarks in 3 categories · methodology

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro41.0%#81 / 119#197 / 259
ARC Challenge82.8%#10 / 51#16 / 77
OpenBookQA82.6%#4 / 21#6 / 42
TriviaQA67.7%#13 / 18#31 / 39
MMLU68.8%#27 / 77#66 / 136

Math

BenchmarkScoreOpen rankAll models
AIME 2024/20251.9%#66 / 74#255 / 270
MATH Level 56.1%#29 / 32#104 / 108

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond26.1%#76 / 83#280 / 291
WinoGrande65.0%#32 / 46#61 / 80

Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.