Alibaba (Qwen) · model
Qwen3.5-35B-A3B benchmark scores
As of 2026-11-21, Qwen3.5-35B-A3B (Alibaba (Qwen)) has measured scores on 28 benchmarks and estimated scores on 103 more.
| Benchmark | Score | Source |
|---|---|---|
| AA-GPQA Diamond | 84.5% | measured |
| AA-HLE | 21.0% | measured |
| AA-SciCode | 43.3% | estimated ± 3.8 pp, high confidence |
| Artificial Analysis Intelligence Index | 19.3% | measured |
| CritPt | 0.9% | measured |
| GDPval-AA | 22.9% | estimated ± 7.8 pp, medium confidence |
| GPQA Diamond | 84.3% | estimated ± 1.2 pp, high confidence |
| HLE | 18.1% | estimated ± 5.1 pp, medium confidence |
| IFBench | 70.5% | estimated ± 6.6 pp, medium confidence |
| AA-LCR | 72.0% | measured |
| LiveCodeBench v6 | 74.6% | measured |
| AA-Omniscience Accuracy | 20.1% | measured |
| SciCode | 42.6% | estimated ± 3.6 pp, high confidence |
| BrowseComp | 61.0% | measured |
| HLE w/ tools | 43.9% | estimated ± 5.5 pp, medium confidence |
| IFEval | 91.9% | measured |
| LongBench v2 | 59.0% | measured |
| VITA-Bench | 32.6% | estimated ± 6.7 pp, low confidence |
| τ²-bench results | 89.2% | measured |
| AA Coding Index | 38.9% | estimated ± 5.3 pp, medium confidence |
| AA-MMMU-Pro | 72.7% | measured |
| APEX-Agents-AA | 11.3% | estimated ± 4.4 pp, high confidence |
| BioMysteryBench (human-difficult) | 15.7% | estimated ± 6.8 pp, low confidence |
| DeepSearchQA | 77.0% | estimated ± 6.8 pp, low confidence |
| SWE-bench Verified | 69.2% | measured |
| BFCL v4 | 66.1% | estimated ± 8.5 pp, low confidence |
| CyberGym | 47.2% | estimated ± 7.3 pp, low confidence |
| JobBench | 11.6% | estimated ± 10.3 pp, low confidence |
| SkillsBench | 33.8% | estimated ± 4.1 pp, low confidence |
| SWE-bench Pro | 46.5% | estimated ± 4.5 pp, high confidence |
| WideResearch | 73.3% | estimated ± 2.1 pp, low confidence |
| AA Agentic Index | 7.0% | estimated ± 6.8 pp, low confidence |
| MMLU-Pro | 85.3% | measured |
| AA-IFBench | 72.5% | measured |
| BBH | 99.0% | estimated ± 2.8 pp, medium confidence |
| GPQA | 84.2% | measured |
| MMLU | 86.9% | estimated ± 0.9 pp, high confidence |
| Vals GPQA Diamond | 83.6% | estimated ± 3.2 pp, high confidence |
| Vals LiveCodeBench | 69.8% | estimated ± 10.0 pp, low confidence |
| Vals MMLU-Pro | 84.6% | estimated ± 2.3 pp, high confidence |
| MMMLU | 86.3% | estimated ± 0.7 pp, medium confidence |
| MMMU | 81.4% | measured |
| AA MMLU-Pro | 84.1% | estimated ± 0.8 pp, medium confidence |
| Gert Labs | 29.0% | measured |
| AA Harvey LAB v1.0 | 85.7% | estimated ± 9.8 pp, low confidence |
| AA Tau3 Banking | 23.5% | estimated ± 3.6 pp, low confidence |
| ARC-AGI-1 | 64.4% | estimated ± 6.7 pp, medium confidence |
| ARC-AGI-2 | 16.3% | estimated ± 10.9 pp, low confidence |
| FrontierCode 1.1 Main | 8.4% | estimated ± 4.2 pp, low confidence |
| MazeBench | 0.0% | estimated ± 6.4 pp, low confidence |
| OfficeQA Pro | 50.3% | estimated ± 6.8 pp, low confidence |
| OSWorld-Verified | 54.5% | measured |
| ParseBench-1 | 56.6% | estimated ± 6.2 pp, medium confidence |
| Terminal-Bench Hard | 41.4% | estimated ± 4.0 pp, medium confidence |
| Vals SWE-bench | 61.6% | estimated ± 6.6 pp, medium confidence |
| Terminal-Bench 2.1 (Vals AI) | 35.1% | estimated ± 6.4 pp, medium confidence |
| CWE-Bench-AA | 41.1% | estimated ± 6.4 pp, medium confidence |
| AA Cyber Index | 30.8% | estimated ± 13.5 pp, low confidence |
| HealthBench Professional (length-adjusted, Artificial Analysis) | 51.7% | estimated ± 4.3 pp, low confidence |
| HealthBench Professional (raw, Artificial Analysis) | 54.8% | estimated ± 1.1 pp, low confidence |
| AA ITBench | 36.0% | estimated ± 2.8 pp, medium confidence |
| Chartography (tools) | 78.8% | estimated ± 4.1 pp, low confidence |
| DeepSWE | 27.7% | estimated ± 12.7 pp, low confidence |
| DRACO | 60.9% | estimated ± 2.9 pp, low confidence |
| HLE w/o tools | 31.3% | estimated ± 4.9 pp, high confidence |
| OSWorld 2.0 | 0.0% | estimated ± 3.6 pp, medium confidence |
| SWE Multilingual | 68.1% | estimated ± 3.2 pp, low confidence |
| Toolathlon-Verified | 45.6% | estimated ± 7.4 pp, medium confidence |
| ARC-AGI-3 | 1.3% | estimated ± 9.8 pp, medium confidence |
| ARC-AGI-3 Provider Adapter | 2.7% | estimated ± 7.3 pp, low confidence |
| Vibe Code Bench | 0.0% | estimated ± 8.4 pp, medium confidence |
| ParseBench-2 | 65.0% | estimated ± 5.0 pp, medium confidence |
| CharXiv | 77.5% | measured |
| CharXiv w/o tools | 79.4% | estimated ± 1.2 pp, high confidence |
| AI2D_TEST | 91.1% | estimated ± 0.9 pp, medium confidence |
| C-Eval | 91.4% | estimated ± 1.0 pp, medium confidence |
| Claw-Eval | 52.9% | estimated ± 6.2 pp, low confidence |
| CountBench | 93.3% | estimated ± 1.3 pp, medium confidence |
| DynaMath | 84.8% | estimated ± 0.8 pp, medium confidence |
| ERQA | 52.9% | estimated ± 3.1 pp, high confidence |
| INCLUDE | 82.5% | estimated ± 1.4 pp, low confidence |
| MathVision | 83.9% | measured |
| MAXIFE | 71.2% | estimated ± 6.5 pp, low confidence |
| MCP Atlas | 49.4% | estimated ± 10.3 pp, low confidence |
| MLVU (M-Avg) | 84.8% | estimated ± 1.5 pp, high confidence |
| MMLU-ProX | 81.0% | measured |
| MMLU-Redux | 93.3% | estimated ± 0.6 pp, high confidence |
| MMMU-Pro | 75.1% | measured |
| NL2Repo | 22.6% | estimated ± 3.4 pp, medium confidence |
| NOVA-63 | 55.7% | estimated ± 1.3 pp, low confidence |
| OmniDocBench 1.5 | 88.8% | estimated ± 2.2 pp, high confidence |
| PolyMath | 39.2% | estimated ± 11.4 pp, low confidence |
| RealWorldQA | 82.4% | estimated ± 1.5 pp, high confidence |
| ScreenSpot Pro | 67.1% | estimated ± 6.8 pp, medium confidence |
| SimpleVQA | 61.4% | estimated ± 7.1 pp, medium confidence |
| SuperGPQA | 63.4% | measured |
| Terminal-Bench 2.0 | 40.5% | measured |
| Toolathlon | 32.6% | estimated ± 4.2 pp, high confidence |
| V* | 92.7% | measured |
| Video-MME (w/o subtitle) | 81.5% | estimated ± 1.2 pp, high confidence |
| Video-MME (with subtitle) | 85.4% | estimated ± 1.9 pp, high confidence |
| VideoMMMU | 84.9% | estimated ± 1.5 pp, medium confidence |
| We-Math | 83.6% | estimated ± 2.1 pp, medium confidence |
| HealthBench Hard | 8.9% | estimated ± 8.6 pp, low confidence |
| MedXpertQA (MM) | 63.5% | estimated ± 2.4 pp, high confidence |
| MMLU-Pro (Arcee) | 80.5% | estimated ± 4.0 pp, medium confidence |
| React Native Evals | 74.8% | estimated ± 2.7 pp, medium confidence |
| ResearchClawBench | 13.5% | estimated ± 2.1 pp, low confidence |
| SWE-Rebench | 53.7% | measured |
| BioMysteryBench (human-solvable) | 81.7% | estimated ± 0.6 pp, low confidence |
| HealthBench (raw) | 28.9% | estimated ± 6.0 pp, low confidence |
| HealthBench (length-adjusted) | 46.4% | estimated ± 3.4 pp, low confidence |
| HealthBench Professional | 56.7% | estimated ± 4.2 pp, low confidence |
| HealthBench Professional (raw) | 40.8% | estimated ± 4.5 pp, medium confidence |
| HLE-Verified | 15.5% | estimated ± 7.1 pp, low confidence |
| STEMBench | 2.3% | estimated ± 0.2 pp, low confidence |
| SimpleQA | 18.2% | estimated ± 4.3 pp, medium confidence |
| LiveCodeBench | 72.4% | estimated ± 14.6 pp, low confidence |
| Agents' Last Exam | 14.2% | estimated ± 11.6 pp, low confidence |
| BabyVision w/ Python | 88.8% | estimated ± 3.8 pp, low confidence |
| BabyVision | 42.3% | estimated ± 3.7 pp, high confidence |
| MMVU | 72.3% | measured |
| PerceptionBench | 50.6% | estimated ± 3.6 pp, low confidence |
| ZeroBench | 15.7% | estimated ± 3.6 pp, low confidence |
| MRCRv2 | 80.1% | estimated ± 11.3 pp, low confidence |
| MMLongBench-Doc | 61.0% | estimated ± 2.6 pp, low confidence |
| OCRBench V2 | 70.3% | estimated ± 3.3 pp, high confidence |
| BioBasicsBench | 33.9% | estimated ± 2.8 pp, low confidence |
| LifeVal | 39.3% | estimated ± 1.6 pp, low confidence |
| MMMU-Pro w/ Python | 76.2% | estimated ± 2.0 pp, high confidence |
| PinchBench | 88.9% | estimated ± 3.6 pp, medium confidence |