Qwen3 235B A22B Thinking 2507 — Benchmarks
Benchmark scores for Qwen3 235B A22B Thinking 2507 aggregated from public leaderboards, with how it ranks among open models. See hardware requirements for what you need to run it.
Overall rank: #63 of 123 open modelscomposite 47.1/100 across 7 benchmarks in 4 categories · methodology
Coding
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| SciCode | 42.4% | #23 / 50 | #99 / 155 |
Knowledge
Math
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| AIME 2024/2025 | 86.7% | #16 / 74 | #74 / 270 |
| FrontierMath | 8.5% | #7 / 12 | #59 / 101 |
Reasoning
| Benchmark | Score | Open rank | All models |
|---|---|---|---|
| GPQA Diamond | 80.0% | #21 / 83 | #112 / 291 |
| CritPt | 0.0% | #47 / 50 | #163 / 167 |
Scores aggregated from public benchmark sources (each linked from the benchmark pages). llmrun does not run these benchmarks.