Llama 2 70B Chat — Benchmarks

Benchmark scores for Llama 2 70B Chat aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Knowledge

BenchmarkScoreRank
MMLU59.9%#21 / 36

Math

BenchmarkScoreRank
GSM8K58.7%#14 / 27

Reasoning

BenchmarkScoreRank
BIG-Bench Hard58.5%#4 / 11

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