QwQ 32B — Benchmarks

Benchmark scores for QwQ 32B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #56 of 123 open modelscomposite 49.9/100 across 4 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
Aider Polyglot20.9%#12 / 18#57 / 69

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro69.1%#37 / 119#105 / 259

Math

BenchmarkScoreOpen rankAll models
AIME 2024/202559.2%#34 / 74#152 / 270

Reasoning

BenchmarkScoreOpen rankAll models
GPQA Diamond65.3%#34 / 83#181 / 291

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