Qwen2.5 Coder 7B — Benchmarks

Benchmark scores for Qwen2.5 Coder 7B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #20 of 123 open modelscomposite 64/100 across 5 benchmarks in 3 categories · methodology

Knowledge

BenchmarkScoreOpen rankAll models
HellaSwag76.8%#20 / 42#39 / 76
ARC Challenge60.9%#23 / 51#34 / 77
MMLU68.0%#32 / 77#73 / 136

Math

BenchmarkScoreOpen rankAll models
GSM8K83.9%#11 / 60#17 / 93

Reasoning

BenchmarkScoreOpen rankAll models
WinoGrande72.9%#21 / 46#44 / 80

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