Llama 7B — Benchmarks

Benchmark scores for Llama 7B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Knowledge

BenchmarkScoreRank
MMLU35.2%#32 / 36
HellaSwag56.2%#17 / 22

Math

BenchmarkScoreRank
GSM8K11.0%#24 / 27

Reasoning

BenchmarkScoreRank
BIG-Bench Hard30.3%#10 / 11

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