Qwen3 235B A22B — Benchmarks

Benchmark scores for Qwen3 235B A22B aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.

Overall rank: #17 of 73 open modelscomposite 59.8/100 across 5 benchmarks in 4 categories · methodology

Coding

BenchmarkScoreOpen rankAll models
Aider Polyglot59.6%#3 / 18#27 / 69

Knowledge

BenchmarkScoreOpen rankAll models
MMLU-Pro68.2%#38 / 119#108 / 259

Math

BenchmarkScoreOpen rankAll models
MATH Level 568.9%#8 / 32#45 / 108

Reasoning

BenchmarkScoreOpen rankAll models
SimpleBench31.0%#11 / 19#68 / 90
GPQA Diamond70.7%#12 / 46#86 / 182

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