Alibaba (Qwen) · model
Qwen3.8 Max benchmark scores
As of 2026-11-22, Qwen3.8 Max (Alibaba (Qwen)) has measured scores on 68 benchmarks and estimated scores on 120 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 91.0% | estimated ± 2.7 pp, high confidence |
| AA-HLE | 42.4% | estimated ± 3.9 pp, high confidence |
| AA-SciCode | 52.0% | estimated ± 3.1 pp, high confidence |
| AIME26 | 99.2% | estimated ± 1.2 pp, low confidence |
| Artificial Analysis Intelligence Index | 34.2% | estimated ± 5.6 pp, medium confidence |
| CritPt | 17.9% | estimated ± 4.0 pp, high confidence |
| GDPval-AA | 52.5% | estimated ± 5.6 pp, medium confidence |
| GPQA Diamond | 92.6% | measured |
| HLE | 43.6% | measured |
| IFBench | 82.8% | measured |
| AA-LCR | 74.3% | estimated ± 3.4 pp, medium confidence |
| LiveCodeBench v6 | 95.8% | estimated ± 1.7 pp, medium confidence |
| AA-Omniscience Accuracy | 45.6% | estimated ± 6.5 pp, medium confidence |
| SciCode | 52.5% | estimated ± 2.9 pp, low confidence |
| Terminal-Bench 2.1 | 86.6% | measured |
| BrowseComp | 82.9% | estimated ± 1.9 pp, high confidence |
| HLE w/ tools | 56.2% | measured |
| IFEval | 95.0% | estimated ± 3.6 pp, high confidence |
| LongBench v2 | 66.3% | measured |
| τ²-bench results | 98.0% | estimated ± 3.1 pp, high confidence |
| AA Coding Index | 72.0% | estimated ± 2.5 pp, high confidence |
| AA-MMMU-Pro | 81.4% | estimated ± 1.5 pp, high confidence |
| APEX-Agents | 37.5% | estimated ± 7.4 pp, low confidence |
| APEX-Agents-AA | 36.3% | estimated ± 2.9 pp, medium confidence |
| DeepSearchQA | 91.3% | estimated ± 7.9 pp, low confidence |
| SWE-bench Verified | 81.6% | estimated ± 2.3 pp, high confidence |
| AutomationBench | 27.3% | measured |
| CyberGym | 78.7% | estimated ± 4.6 pp, medium confidence |
| JobBench | 53.4% | measured |
| SkillsBench | 70.2% | measured |
| SWE-bench Pro | 67.7% | measured |
| WideResearch | 81.9% | measured |
| AA Agentic Index | 48.5% | estimated ± 3.3 pp, medium confidence |
| MMLU-Pro | 85.7% | estimated ± 2.5 pp, low confidence |
| AA-IFBench | 80.4% | estimated ± 7.4 pp, low confidence |
| FailureSensorIQ | 67.9% | estimated ± 1.7 pp, high confidence |
| GPQA | 92.6% | measured |
| MMLU | 95.2% | estimated ± 1.2 pp, medium confidence |
| Vals GPQA Diamond | 93.7% | measured |
| Vals LiveCodeBench | 87.9% | measured |
| Vals MMLU-Pro | 88.6% | measured |
| MMMLU | 90.7% | estimated ± 1.1 pp, low confidence |
| MMMU | 88.2% | estimated ± 1.7 pp, medium confidence |
| Chess Puzzles | 46.0% | measured |
| AA MMLU-Pro | 88.4% | estimated ± 1.1 pp, high confidence |
| Gert Labs | 55.7% | measured |
| AA-AnalystAgent | 42.0% | estimated ± 4.6 pp, high confidence |
| AA EnterpriseOps-Gym | 46.3% | estimated ± 1.2 pp, medium confidence |
| AA Harvey LAB v1.0 | 89.7% | estimated ± 1.3 pp, medium confidence |
| AA Tau3 Banking | 41.8% | estimated ± 3.3 pp, medium confidence |
| AA Terminal-Bench 2.1 | 82.0% | estimated ± 2.5 pp, high confidence |
| ApprenticeBench | 17.0% | estimated ± 6.0 pp, low confidence |
| ARC-AGI-1 | 93.0% | estimated ± 2.5 pp, high confidence |
| ARC-AGI-2 | 71.2% | estimated ± 9.7 pp, medium confidence |
| BabaIsBench | 58.0% | measured |
| CursorBench 3.1 | 59.6% | estimated ± 5.6 pp, low confidence |
| CursorBench 3.2 | 64.2% | estimated ± 3.3 pp, medium confidence |
| EBR-bench | 29.0% | estimated ± 10.4 pp, low confidence |
| Terminal-Bench 3.0 | 10.9% | estimated ± 6.9 pp, low confidence |
| FrontierCode 1.1 Main | 42.3% | estimated ± 3.5 pp, low confidence |
| FrontierMath v2 (Tier 4) | 46.3% | measured |
| FrontierMath v2 (Tiers 1-3) | 74.7% | measured |
| FrontierSWE v2 | 15.8% | measured |
| Furniture Assembly | 36.0% | estimated ± 9.3 pp, medium confidence |
| MaxxBench | 64.8% | estimated ± 5.7 pp, medium confidence |
| MazeBench | 0.0% | measured |
| MirrorCode | 52.2% | measured |
| Mystery Game Puzzles | 38.0% | measured |
| NYT Connections | 88.3% | measured |
| OSWorld-Verified | 86.1% | measured |
| ParseBench-1 | 71.2% | estimated ± 1.0 pp, low confidence |
| Terminal-Bench 4.1 | 23.5% | estimated ± 10.9 pp, low confidence |
| Terminal-Bench Hard | 51.7% | estimated ± 3.9 pp, high confidence |
| Vals SWE-bench | 85.6% | measured |
| Terminal-Bench 2.1 (Vals AI) | 67.4% | measured |
| VulcanBench v3 | 81.2% | measured |
| CWE-Bench-AA | 59.7% | estimated ± 4.0 pp, medium confidence |
| AA Cyber Index | 32.8% | estimated ± 9.4 pp, low confidence |
| DeepsecBench-AA | 17.5% | estimated ± 5.2 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 57.9% | estimated ± 4.4 pp, low confidence |
| HealthBench Professional (raw, Artificial Analysis) | 64.9% | estimated ± 2.6 pp, low confidence |
| AA ITBench | 46.1% | estimated ± 4.6 pp, medium confidence |
| Bug Hunt Bench | 25.7% | measured |
| CursorBench 4.0 | 36.7% | estimated ± 2.6 pp, medium confidence |
| CWE-bench v1 | 57.4% | estimated ± 5.1 pp, medium confidence |
| DeepSWE | 56.6% | measured |
| DRACO | 86.9% | estimated ± 0.7 pp, low confidence |
| FrontierCode 2.0 | 24.3% | estimated ± 4.8 pp, medium confidence |
| FrontierPhysics | 11.8% | estimated ± 4.7 pp, high confidence |
| HLE w/o tools | 43.6% | measured |
| OSWorld 2.0 | 19.4% | measured |
| OSWorld 2.1 | 69.8% | estimated ± 3.6 pp, high confidence |
| OSWorld 2.1 Offline | 63.9% | estimated ± 5.5 pp, low confidence |
| PostTrainBench v1.1 | 28.6% | estimated ± 3.7 pp, high confidence |
| SpireClimb | 5.8% | estimated ± 7.2 pp, medium confidence |
| StealthMark | 62.4% | measured |
| Toolathlon-Verified | 72.5% | measured |
| VulcanBench Frontier v4 | 77.1% | estimated ± 5.0 pp, high confidence |
| AA AutomationBench | 41.9% | estimated ± 7.9 pp, low confidence |
| GDP.pdf | 16.8% | estimated ± 4.1 pp, medium confidence |
| AA Terminal-Bench 4.0 | 34.2% | estimated ± 11.0 pp, low confidence |
| ARC-AGI-3 | 5.3% | estimated ± 7.2 pp, medium confidence |
| cursorBench41 | 38.4% | estimated ± 4.3 pp, medium confidence |
| CWE-bench | 45.4% | measured |
| MKA | 38.5% | estimated ± 3.2 pp, high confidence |
| SkillsBench official leaderboard (BenchFlow) | 56.6% | estimated ± 6.1 pp, medium confidence |
| BinaryAudit | 35.8% | estimated ± 12.1 pp, low confidence |
| CompileBench | 100.0% | estimated ± 10.0 pp, low confidence |
| OTelBench | 22.5% | estimated ± 6.6 pp, medium confidence |
| FACTS Multimodal | 51.5% | measured |
| Hermes Bench | 70.5% | measured |
| Hermes Index | 36.2% | measured |
| ParseBench-2 | 75.6% | estimated ± 5.0 pp, medium confidence |
| CharXiv | 93.5% | measured |
| CharXiv w/o tools | 88.4% | measured |
| ExploitBench | 59.8% | estimated ± 7.2 pp, medium confidence |
| ExploitGym | 11.9% | estimated ± 6.3 pp, medium confidence |
| AI2D_TEST | 93.5% | estimated ± 0.9 pp, low confidence |
| C-Eval | 93.2% | estimated ± 1.2 pp, medium confidence |
| Claw-Eval | 72.8% | estimated ± 4.5 pp, medium confidence |
| CountBench | 97.3% | estimated ± 1.3 pp, low confidence |
| DynaMath | 88.5% | estimated ± 0.8 pp, low confidence |
| ERQA | 77.8% | measured |
| HMMT Feb 2026 | 100.0% | estimated ± 1.5 pp, low confidence |
| HMMT Nov 2025 | 95.6% | estimated ± 1.5 pp, low confidence |
| MathVision | 95.2% | measured |
| MCP Atlas | 85.5% | estimated ± 6.6 pp, low confidence |
| MCP-Tasks | 84.2% | estimated ± 6.0 pp, low confidence |
| MLVU (M-Avg) | 90.8% | measured |
| MMAnswerBench | 92.9% | estimated ± 2.4 pp, low confidence |
| MMLU-Redux | 95.1% | estimated ± 1.7 pp, medium confidence |
| MMMU-Pro | 82.3% | measured |
| NL2Repo | 55.9% | measured |
| OmniDocBench 1.5 | 92.1% | measured |
| PostTrainBench v1 | 31.3% | estimated ± 3.6 pp, high confidence |
| RealWorldQA | 88.0% | measured |
| ScreenSpot Pro | 84.5% | measured |
| SimpleVQA | 75.0% | measured |
| SuperGPQA | 83.2% | estimated ± 7.2 pp, medium confidence |
| Terminal-Bench 2.0 | 71.3% | estimated ± 5.8 pp, medium confidence |
| Toolathlon | 49.4% | estimated ± 5.2 pp, low confidence |
| V* | 96.1% | estimated ± 6.5 pp, low confidence |
| Video-MME (w/o subtitle) | 86.7% | estimated ± 1.3 pp, medium confidence |
| Video-MME (with subtitle) | 90.4% | measured |
| VideoMMMU | 88.7% | measured |
| We-Math | 87.5% | estimated ± 2.1 pp, low confidence |
| AA LiveCodeBench | 95.0% | estimated ± 2.6 pp, medium confidence |
| HealthBench Hard | 33.4% | estimated ± 5.1 pp, low confidence |
| LiveCodeBench Pro | 89.4% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 80.4% | measured |
| MedXpertQA (Text) | 62.4% | estimated ± 4.5 pp, medium confidence |
| MMLU-Pro (Arcee) | 87.7% | estimated ± 3.4 pp, low confidence |
| React Native Evals | 83.9% | estimated ± 4.7 pp, low confidence |
| ResearchClawBench | 20.7% | estimated ± 1.9 pp, medium confidence |
| PostTrainBench | 33.1% | estimated ± 2.4 pp, medium confidence |
| DAYJOB | 13.6% | estimated ± 3.2 pp, high confidence |
| DAYJOB Healthcare | 0.7% | estimated ± 4.2 pp, high confidence |
| Finance Agent v2 | 56.4% | estimated ± 2.4 pp, medium confidence |
| FrontierCode 1.1 Extended | 58.3% | estimated ± 3.0 pp, medium confidence |
| HealthBench Professional | 52.7% | estimated ± 4.2 pp, medium confidence |
| HLE-Verified | 54.3% | estimated ± 4.6 pp, medium confidence |
| LABBench2 | 82.3% | estimated ± 1.8 pp, low confidence |
| ScarfBench | 63.7% | measured |
| BlenderWorld (text-only) | 73.2% | estimated ± 2.8 pp, medium confidence |
| Cookie-Bench | 83.1% | estimated ± 1.3 pp, medium confidence |
| SimpleQA | 38.6% | estimated ± 13.4 pp, low confidence |
| Agents' Last Exam | 52.4% | measured |
| BabyVision w/ Python | 91.3% | measured |
| OpenHarmony Bench v1.0 | 60.8% | measured |
| OpenHarmony Bench v1.2 | 67.8% | measured |
| SEC-Bench Pro | 82.8% | estimated ± 5.6 pp, medium confidence |
| ZeroBench w/ Python | 49.0% | measured |
| BabyVision | 82.0% | measured |
| MMVU | 82.4% | measured |
| PerceptionBench | 63.5% | measured |
| ZeroBench | 24.0% | measured |
| MRCRv2 | 92.9% | measured |
| LVBench | 81.8% | measured |
| CC-OCR | 79.6% | measured |
| RefCOCO (avg) | 86.7% | estimated ± 1.8 pp, medium confidence |
| OCRBench V2 | 74.2% | measured |
| SWE-Marathon | 46.5% | estimated ± 2.1 pp, medium confidence |
| BioBasicsBench | 31.2% | estimated ± 2.2 pp, medium confidence |
| LifeVal | 75.2% | estimated ± 3.8 pp, medium confidence |
| MMMU-Pro w/ Python | 83.2% | estimated ± 2.6 pp, high confidence |
| ExploitBench Internal Port | 1.0% | estimated ± 1.4 pp, medium confidence |
| MathVision w/ Python | 97.7% | measured |
| SWE-Atlas Codebase QnA | 53.7% | estimated ± 4.8 pp, high confidence |