Llama 3.1 70B Instruct — Benchmarks

Benchmark scores for Llama 3.1 70B Instruct aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Knowledge

BenchmarkScoreRank
MMLU80.1%#5 / 36

Math

BenchmarkScoreRank
AIME 2024/20253.6%#17 / 22
MATH Level 536.7%#13 / 23

Reasoning

BenchmarkScoreRank
GPQA Diamond44.2%#18 / 28

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