Alibaba (Qwen) · model
Qwen4 Max benchmark scores
As of 2026-11-18, Qwen4 Max (Alibaba (Qwen)) has measured scores on 15 benchmarks and estimated scores on 76 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 92.8% | estimated ± 2.1 pp, medium confidence |
| AA-HLE | 50.5% | estimated ± 4.6 pp, high confidence |
| AA-SciCode | 59.0% | estimated ± 3.9 pp, medium confidence |
| Artificial Analysis Intelligence Index | 42.9% | measured |
| CritPt | 21.3% | estimated ± 5.3 pp, medium confidence |
| GDPval-AA | 61.7% | estimated ± 8.6 pp, medium confidence |
| GPQA Diamond | 94.1% | estimated ± 1.2 pp, medium confidence |
| HLE | 51.7% | measured |
| AA-Omniscience Accuracy | 43.7% | measured |
| SciCode | 62.2% | estimated ± 7.6 pp, low confidence |
| Terminal-Bench 2.1 | 89.1% | estimated ± 5.4 pp, low confidence |
| BrowseComp | 93.0% | estimated ± 4.2 pp, medium confidence |
| HLE w/ tools | 65.5% | estimated ± 6.8 pp, low confidence |
| τ²-bench results | 95.7% | estimated ± 7.0 pp, low confidence |
| AA Coding Index | 77.9% | estimated ± 2.5 pp, medium confidence |
| APEX-Agents-AA | 59.1% | measured |
| BioMysteryBench (human-difficult) | 44.4% | estimated ± 7.8 pp, low confidence |
| DeepSearchQA | 100.0% | estimated ± 7.9 pp, low confidence |
| SWE-bench Verified | 90.8% | estimated ± 5.4 pp, low confidence |
| AutomationBench | 47.3% | estimated ± 6.1 pp, low confidence |
| JobBench | 81.7% | estimated ± 8.2 pp, low confidence |
| SWE-bench Pro | 92.2% | estimated ± 7.7 pp, low confidence |
| AA Agentic Index | 49.1% | estimated ± 12.2 pp, low confidence |
| MMLU-Pro | 85.9% | estimated ± 2.5 pp, low confidence |
| GPQA | 94.1% | estimated ± 1.2 pp, medium confidence |
| MMLU | 90.5% | estimated ± 0.9 pp, medium confidence |
| Vals GPQA Diamond | 93.5% | estimated ± 1.2 pp, medium confidence |
| Vals LiveCodeBench | 77.4% | estimated ± 6.1 pp, low confidence |
| Vals MMLU-Pro | 90.2% | estimated ± 1.8 pp, high confidence |
| MMMLU | 94.6% | estimated ± 3.0 pp, low confidence |
| AA MMLU-Pro | 87.7% | estimated ± 0.8 pp, high confidence |
| Gert Labs | 74.8% | estimated ± 10.4 pp, low confidence |
| AA-AnalystAgent | 46.3% | measured |
| AA EnterpriseOps-Gym | 48.5% | measured |
| ARC-AGI-1 | 96.2% | estimated ± 3.9 pp, high confidence |
| ARC-AGI-2 | 81.7% | estimated ± 13.0 pp, low confidence |
| CursorBench 3.2 | 31.9% | estimated ± 3.6 pp, low confidence |
| FrontierCode 1.1 Main | 52.8% | estimated ± 3.5 pp, low confidence |
| MazeBench | 3.4% | estimated ± 6.2 pp, low confidence |
| Terminal-Bench 4.1 | 48.3% | estimated ± 5.0 pp, medium confidence |
| Vals SWE-bench | 85.5% | estimated ± 10.6 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 78.5% | estimated ± 6.9 pp, low confidence |
| CWE-Bench-AA | 64.1% | estimated ± 9.0 pp, medium confidence |
| DeepsecBench-AA | 39.4% | estimated ± 7.5 pp, medium confidence |
| Harvey LAB-AA v1.1 | 7.5% | estimated ± 3.5 pp, medium confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 58.9% | estimated ± 4.3 pp, medium confidence |
| HealthBench Professional (raw, Artificial Analysis) | 67.2% | estimated ± 1.1 pp, medium confidence |
| AA ITBench | 47.8% | estimated ± 6.3 pp, low confidence |
| Bug Hunt Bench | 38.2% | estimated ± 10.4 pp, low confidence |
| Chartography (no tools) | 54.2% | measured |
| Chartography (tools) | 87.4% | estimated ± 2.3 pp, medium confidence |
| CursorBench 4.0 | 48.0% | estimated ± 4.7 pp, high confidence |
| DeepSWE | 75.7% | measured |
| HLE w/o tools | 51.7% | measured |
| OSWorld 2.1 Offline | 59.6% | estimated ± 7.9 pp, low confidence |
| PostTrainBench v1.1 | 44.2% | estimated ± 4.5 pp, high confidence |
| ProgramBench | 31.5% | measured |
| SWE-bench Multimodal | 56.1% | estimated ± 8.3 pp, low confidence |
| Terminal-Bench 4.0 | 48.5% | measured |
| Terminal-Bench-Science 0.1 | 37.1% | estimated ± 14.9 pp, low confidence |
| Toolathlon-Verified | 76.6% | estimated ± 2.8 pp, low confidence |
| VulcanBench Frontier v4 | 88.5% | estimated ± 1.2 pp, medium confidence |
| AA AutomationBench | 72.6% | measured |
| GDP.pdf | 26.2% | measured |
| AA Terminal-Bench 4.0 | 48.5% | measured |
| ARC-AGI-3 | 13.6% | estimated ± 7.2 pp, medium confidence |
| ARC-AGI-3 Provider Adapter | 8.1% | estimated ± 11.0 pp, low confidence |
| CWE-bench | 54.5% | estimated ± 6.2 pp, low confidence |
| MKA | 49.3% | estimated ± 5.3 pp, medium confidence |
| Hermes Bench | 76.4% | estimated ± 1.8 pp, medium confidence |
| Hermes Index | 63.3% | estimated ± 2.7 pp, medium confidence |
| ExploitGym | 33.1% | estimated ± 9.8 pp, low confidence |
| C-Eval | 98.4% | estimated ± 1.0 pp, low confidence |
| MCP Atlas | 97.8% | estimated ± 9.1 pp, low confidence |
| MMLU-Redux | 96.4% | estimated ± 0.8 pp, low confidence |
| NL2Repo | 63.3% | estimated ± 6.0 pp, medium confidence |
| SuperGPQA | 90.8% | estimated ± 7.2 pp, medium confidence |
| Toolathlon | 64.7% | estimated ± 6.7 pp, low confidence |
| MedXpertQA (Text) | 68.5% | estimated ± 8.9 pp, low confidence |
| ResearchClawBench | 20.3% | estimated ± 2.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.3% | estimated ± 2.2 pp, high confidence |
| HealthBench Professional | 60.0% | measured |
| HLE-Verified | 38.8% | estimated ± 7.1 pp, low confidence |
| ScarfBench | 69.7% | estimated ± 3.8 pp, medium confidence |
| STEMBench | 2.5% | estimated ± 0.2 pp, medium confidence |
| SimpleQA | 50.7% | estimated ± 4.3 pp, low confidence |
| Agents' Last Exam | 47.0% | estimated ± 12.9 pp, low confidence |
| SWE-Marathon | 46.1% | estimated ± 12.9 pp, low confidence |
| BioBasicsBench | 34.1% | estimated ± 3.8 pp, medium confidence |
| LifeVal | 77.8% | estimated ± 2.3 pp, low confidence |
| SWE-Atlas Codebase QnA | 61.7% | estimated ± 4.8 pp, medium confidence |