Llama 2 13B HF — Benchmarks

Benchmark scores for Llama 2 13B HF aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #42 of 73 open modelscomposite 41.8/100 across 5 benchmarks in 3 categories · methodology

Knowledge

BenchmarkScoreOpen rankAll models
HellaSwag80.7%#14 / 42#27 / 76
MMLU-Pro25.3%#103 / 119#236 / 259
MMLU55.6%#52 / 76#101 / 136

Math

BenchmarkScoreOpen rankAll models
GSM8K34.3%#40 / 59#59 / 93

Reasoning

BenchmarkScoreOpen rankAll models
BIG-Bench Hard47.0%#22 / 37#30 / 50

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