Alibaba (Qwen) · model
Qwen3 235B-A22B (Reasoning) benchmark scores
As of 2026-10-29, Qwen3 235B-A22B (Reasoning) (Alibaba (Qwen)) has measured scores on 9 benchmarks and estimated scores on 22 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 69.1% | estimated ± 6.3 pp, medium confidence |
| AA-HLE | 7.8% | estimated ± 5.0 pp, high confidence |
| AIME26 | 82.0% | estimated ± 4.6 pp, medium confidence |
| CritPt | 0.0% | estimated ± 0.8 pp, low confidence |
| GPQA Diamond | 71.3% | estimated ± 1.2 pp, high confidence |
| HLE | 21.2% | estimated ± 2.3 pp, low confidence |
| IFBench | 51.5% | estimated ± 9.2 pp, medium confidence |
| IFEval | 83.4% | measured |
| MMLU-Pro | 84.5% | estimated ± 1.1 pp, high confidence |
| AA-IFBench | 30.5% | estimated ± 8.8 pp, medium confidence |
| AIME 2024 | 85.7% | measured |
| Artificial Analysis Intelligence Index | 13.7% | estimated ± 3.8 pp, medium confidence |
| GPQA | 71.1% | measured |
| MMLU | 87.9% | estimated ± 4.3 pp, high confidence |
| AA-Omniscience Accuracy | 17.1% | estimated ± 5.4 pp, low confidence |
| MATH-500 | 98.0% | measured |
| MMMLU | 84.3% | measured |
| AA MMLU-Pro | 85.0% | estimated ± 1.0 pp, low confidence |
| AIME 2025 | 81.5% | measured |
| ARC-AGI-1 | 18.0% | estimated ± 7.0 pp, low confidence |
| ARC-AGI-2 | 0.0% | estimated ± 11.6 pp, low confidence |
| Vals GPQA Diamond | 57.2% | estimated ± 4.8 pp, medium confidence |
| Vals MMLU-Pro | 28.7% | estimated ± 1.2 pp, medium confidence |
| HLE w/o tools | 11.7% | estimated ± 5.2 pp, medium confidence |
| SWE Multilingual | 73.2% | estimated ± 3.7 pp, low confidence |
| C-Eval | 89.6% | measured |
| HMMT Feb 2025 | 73.2% | estimated ± 3.1 pp, high confidence |
| INCLUDE | 78.7% | measured |
| MMLU-Redux | 92.7% | measured |
| SuperGPQA | 64.4% | estimated ± 2.0 pp, low confidence |
| SimpleQA | 31.7% | estimated ± 14.6 pp, low confidence |