Alibaba (Qwen) · model
Qwen3-VL-235B-A22B Instruct benchmark scores
As of 2026-11-19, Qwen3-VL-235B-A22B Instruct (Alibaba (Qwen)) has measured scores on 8 benchmarks and estimated scores on 20 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-MMMU-Pro | 67.4% | estimated ± 1.5 pp, high confidence |
| MMMU | 78.7% | measured |
| ParseBench-1 | 50.0% | estimated ± 5.2 pp, low confidence |
| ParseBench-2 | 55.7% | estimated ± 5.0 pp, medium confidence |
| CharXiv | 72.5% | estimated ± 3.3 pp, high confidence |
| CharXiv w/o tools | 74.2% | estimated ± 2.1 pp, high confidence |
| AI2D_TEST | 87.8% | estimated ± 1.4 pp, medium confidence |
| CountBench | 97.4% | estimated ± 1.6 pp, low confidence |
| DynaMath | 79.1% | estimated ± 1.0 pp, medium confidence |
| ERQA | 53.8% | estimated ± 4.2 pp, high confidence |
| MathVision | 76.0% | estimated ± 2.9 pp, high confidence |
| MLVU (M-Avg) | 81.5% | estimated ± 2.2 pp, high confidence |
| MMMU-Pro | 68.1% | measured |
| OmniDocBench 1.5 | 90.7% | estimated ± 1.5 pp, low confidence |
| RealWorldQA | 79.3% | measured |
| ScreenSpot Pro | 62.0% | measured |
| SimpleVQA | 59.2% | estimated ± 7.1 pp, medium confidence |
| V* | 90.9% | estimated ± 6.5 pp, low confidence |
| Video-MME (w/o subtitle) | 79.2% | measured |
| Video-MME (with subtitle) | 82.0% | estimated ± 3.0 pp, high confidence |
| VideoMMMU | 74.7% | measured |
| We-Math | 75.6% | estimated ± 5.5 pp, low confidence |
| MedXpertQA (MM) | 46.1% | estimated ± 2.4 pp, medium confidence |
| MMVU | 70.9% | estimated ± 2.7 pp, medium confidence |
| CC-OCR | 82.2% | measured |
| RefCOCO (avg) | 93.5% | estimated ± 1.8 pp, low confidence |
| OCRBench V2 | 67.1% | measured |
| MMMU-Pro w/ Python | 74.1% | estimated ± 2.6 pp, high confidence |