Alibaba (Qwen) · model
Qwen3.6-35B-A3B benchmark scores
As of 2026-10-07, Qwen3.6-35B-A3B (Alibaba (Qwen)) has measured scores on 47 benchmarks and estimated scores on 83 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 84.1% | measured |
| AA-HLE | 22.2% | measured |
| AA-SciCode | 36.6% | measured |
| AIME26 | 92.7% | measured |
| Artificial Analysis Intelligence Index | 18.2% | measured |
| CritPt | 0.3% | measured |
| GDPval-AA | 19.8% | measured |
| GPQA Diamond | 86.0% | estimated ± 0.4 pp, high confidence |
| HLE | 21.4% | measured |
| IFBench | 64.1% | estimated ± 6.9 pp, medium confidence |
| AA-LCR | 71.7% | measured |
| LiveCodeBench v6 | 84.6% | estimated ± 2.1 pp, high confidence |
| AA-Omniscience Accuracy | 18.8% | measured |
| SciCode | 36.7% | estimated ± 3.2 pp, high confidence |
| BrowseComp | 58.9% | estimated ± 4.4 pp, medium confidence |
| HLE w/ tools | 30.6% | estimated ± 2.4 pp, medium confidence |
| IFEval | 93.1% | estimated ± 1.3 pp, high confidence |
| LongBench v2 | 61.5% | estimated ± 5.9 pp, medium confidence |
| VITA-Bench | 35.6% | measured |
| τ²-bench results | 95.3% | measured |
| AA Coding Index | 41.9% | measured |
| AA-MMMU-Pro | 75.0% | measured |
| APEX-Agents | 20.1% | estimated ± 0.6 pp, medium confidence |
| APEX-Agents-AA | 28.3% | estimated ± 5.7 pp, medium confidence |
| BioMysteryBench (human-difficult) | 25.9% | estimated ± 5.9 pp, low confidence |
| DeepSearchQA | 80.0% | estimated ± 4.5 pp, high confidence |
| SWE-bench Verified | 73.4% | measured |
| AutomationBench | 28.4% | estimated ± 9.7 pp, low confidence |
| BFCL v4 | 72.7% | estimated ± 2.6 pp, high confidence |
| CyberGym | 57.1% | estimated ± 5.3 pp, low confidence |
| JobBench | 0.0% | estimated ± 5.1 pp, low confidence |
| SkillsBench | 18.0% | estimated ± 5.6 pp, low confidence |
| SWE-bench Pro | 49.5% | measured |
| τ³-bench results | 67.2% | measured |
| WideResearch | 60.1% | measured |
| AA Agentic Index | 15.0% | measured |
| MMLU-Pro | 85.2% | measured |
| FrontierMath v2 (Tier 4) | 0.0% | estimated ± 3.7 pp, low confidence |
| FrontierMath v2 (Tiers 1-3) | 24.4% | estimated ± 8.4 pp, low confidence |
| GPQA | 86.0% | measured |
| AA-IFBench | 64.4% | measured |
| Gert Labs | 42.7% | measured |
| AA-AnalystAgent | 13.4% | estimated ± 12.0 pp, low confidence |
| AA EnterpriseOps-Gym | 47.1% | estimated ± 3.6 pp, medium confidence |
| AA Harvey LAB | 79.7% | estimated ± 1.2 pp, low confidence |
| ApprenticeBench | 0.0% | estimated ± 3.1 pp, low confidence |
| ARC-AGI-1 | 37.9% | estimated ± 5.4 pp, low confidence |
| ARC-AGI-2 | 27.2% | estimated ± 11.6 pp, low confidence |
| CursorBench 3.1 | 11.0% | estimated ± 3.4 pp, low confidence |
| CursorBench 3.2 | 39.1% | estimated ± 3.6 pp, medium confidence |
| FrontierCode 1.1 Main | 7.7% | estimated ± 2.7 pp, low confidence |
| FrontierSWE v2 | 0.0% | estimated ± 9.0 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 7.2 pp, low confidence |
| OSWorld-Verified | 55.5% | estimated ± 5.9 pp, medium confidence |
| Vals GPQA Diamond | 80.3% | estimated ± 4.8 pp, high confidence |
| Vals LiveCodeBench | 76.5% | estimated ± 4.7 pp, medium confidence |
| Vals MMLU-Pro | 83.2% | estimated ± 1.2 pp, high confidence |
| Vals SWE-bench | 59.7% | estimated ± 5.4 pp, low confidence |
| Terminal-Bench 2.1 (Vals AI) | 45.0% | estimated ± 6.3 pp, medium confidence |
| AA AutomationBench | 3.8% | estimated ± 7.0 pp, medium confidence |
| GDP.pdf | 9.4% | estimated ± 5.4 pp, medium confidence |
| AA ITBench | 32.7% | estimated ± 3.6 pp, low confidence |
| AA Tau3 Banking | 22.0% | estimated ± 5.8 pp, medium confidence |
| CursorBench 4.0 | 11.9% | estimated ± 1.6 pp, medium confidence |
| CWE-bench v1 | 29.2% | estimated ± 6.9 pp, low confidence |
| DeepSWE | 5.4% | estimated ± 6.4 pp, low confidence |
| HLE w/o tools | 22.9% | estimated ± 4.9 pp, high confidence |
| OSWorld 2.0 | 0.0% | estimated ± 4.5 pp, low confidence |
| PostTrainBench v1.1 | 19.7% | estimated ± 4.7 pp, medium confidence |
| SWE Multilingual | 67.2% | measured |
| Toolathlon-Verified | 40.3% | estimated ± 2.3 pp, low confidence |
| Vibe Code Bench | 1.4% | estimated ± 8.1 pp, medium confidence |
| CharXiv | 78.0% | measured |
| CharXiv w/o tools | 79.3% | estimated ± 1.8 pp, medium confidence |
| ExploitGym | 0.0% | estimated ± 12.2 pp, low confidence |
| C-Eval | 90.0% | measured |
| Claw-Eval | 68.7% | measured |
| DeepPlanning | 25.9% | measured |
| HMMT Feb 2025 | 90.7% | measured |
| HMMT Feb 2026 | 83.6% | measured |
| HMMT Nov 2025 | 89.1% | measured |
| MathVision | 85.0% | estimated ± 2.6 pp, high confidence |
| MCP Atlas | 62.8% | measured |
| MCP-Tasks | 76.6% | estimated ± 0.6 pp, low confidence |
| MMAnswerBench | 78.9% | measured |
| MMLU-ProX | 81.7% | estimated ± 1.7 pp, low confidence |
| MMLU-Redux | 93.1% | estimated ± 0.7 pp, high confidence |
| MMMU-Pro | 75.3% | measured |
| NL2Repo | 29.4% | measured |
| NOVA-63 | 53.6% | estimated ± 1.6 pp, low confidence |
| QwenClawBench | 52.6% | measured |
| ScreenSpot Pro | 85.0% | estimated ± 2.5 pp, low confidence |
| SuperGPQA | 64.7% | measured |
| Terminal-Bench 2.0 | 51.5% | measured |
| Toolathlon | 26.9% | measured |
| V* | 93.3% | estimated ± 1.4 pp, medium confidence |
| VideoMMMU | 83.7% | measured |
| ERQA | 66.7% | estimated ± 3.2 pp, medium confidence |
| HealthBench Hard | 0.0% | estimated ± 5.1 pp, low confidence |
| LiveCodeBench Pro | 74.9% | estimated ± 6.0 pp, low confidence |
| MedXpertQA (MM) | 64.0% | estimated ± 2.4 pp, high confidence |
| MMLU-Pro (Arcee) | 78.3% | estimated ± 4.0 pp, low confidence |
| React Native Evals | 74.6% | estimated ± 3.5 pp, high confidence |
| ResearchClawBench | 18.6% | estimated ± 1.3 pp, high confidence |
| SWE-Rebench | 57.0% | estimated ± 3.6 pp, medium confidence |
| ARC-AGI-3 | 0.0% | estimated ± 2.1 pp, medium confidence |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| DRACO | 69.7% | estimated ± 3.0 pp, low confidence |
| FrontierCode 1.1 Extended | 52.1% | estimated ± 3.0 pp, low confidence |
| HealthBench (raw) | 30.6% | estimated ± 6.4 pp, low confidence |
| HealthBench (length-adjusted) | 53.9% | estimated ± 3.7 pp, low confidence |
| HealthBench Professional | 42.4% | estimated ± 4.8 pp, low confidence |
| HealthBench Professional (raw) | 42.4% | estimated ± 5.0 pp, low confidence |
| HLE-Verified | 15.3% | estimated ± 2.0 pp, low confidence |
| LABBench2 | 68.3% | estimated ± 1.8 pp, low confidence |
| AIME 2025 | 93.5% | estimated ± 1.5 pp, medium confidence |
| MMMU | 81.7% | measured |
| LiveCodeBench | 80.4% | measured |
| OpenHarmony Bench | 51.0% | estimated ± 2.4 pp, medium confidence |
| IMOAnswerBench | 80.1% | estimated ± 5.9 pp, medium confidence |
| SimpleVQA | 58.9% | measured |
| ZeroBench | 44.2% | estimated ± 6.4 pp, low confidence |
| MMLU | 85.4% | estimated ± 1.1 pp, medium confidence |
| MMMU-Pro w/ Python | 76.8% | estimated ± 0.6 pp, high confidence |
| RefCOCO (avg) | 92.0% | measured |
| MATH-500 | 97.3% | estimated ± 3.1 pp, medium confidence |
| MathVision w/ Python | 76.8% | estimated ± 0.9 pp, low confidence |
| RealWorldQA | 85.3% | measured |
| Video-MME (with subtitle) | 86.6% | measured |
| OmniDocBench 1.5 | 89.9% | measured |