- Terminal Bench 2.1 86.6
- SWE-bench Pro 67.7
- DeepSWE 1.1 56.6
- NL2Repo-Bench 55.9
- FrontierSWE 73.5
- MLS-Bench-Lite 41
- PaperBench 93
- AndroidBench 75.1
- QwenSWEBench 80.7
- QwenQoderBench 58.4
- QwenReactBench 1724
- QwenSVGBench 1713
- CoWorkBench 74.8
- WorkSpaceBench 67.7
- JobBench 53.4
- SkillsBench 70.2
- Agents' Last Exam (Pass / Score) 27.0 / 52.4 27
- Automation-Bench (Pass@1) 27.3
- Toolathlon Verified (Pass@1) 72.5
- WideSearch 81.9
- Best of HLE w/ tools / HLE 56.2
- GPQA Diamond 92.6
- IFBench 82.8
- $OneMillion-Bench (expert score) 52.5
- HealthBench 60.2
- PLawBench 73.2
- PRBench-Legal 57.6
- PRBench-Finance 58.3
- MRCR v2 256K (8-needle) 92.9
- LongBench v2 66.3
- MMMU-Pro 82.3
- MathVision (without CI / with CI) 95.2 / 97.7 97.7
- BabyVision (without CI / with CI) 82.0 / 91.3 91.3
- HLE-VL (w/ Tools) 52.2
- Best of ZeroBench (Pass@5) (without CI / with CI) 24.0 / 49.0 / ZeroBench-Sub 49
- LogicVista 91.9
- HiPhO 90
- PhyX 83.5
- SLAKE 90.8
- MedXpertQA-MM 80.4
- PMC-VQA 66.2
- OSWorld-Verified 86.1
- OSWorld 2.0 (binary / partial) 19.4 / 46.7 19.4
- ScreenSpot Pro 84.5
- WebArena-Verified 66.8
- AndroidWorld 85.3
- MobileWorld 77.8
- ClawEval-MM (Pass@3 / avg) 77.2 / 74.8 74.8
- Vision2Web 69
- QwenBlenderBench 69.9
- Parametric CAD Bench 91.5
- RecreationBench 51.7
- PresentBench 79.6
- CharXiv (RQ) (without CI / with CI) 88.4 / 93.5 93.5
- OmniDocBench 1.5 92.1
- OCR-Bench-V2 (EN/ZH) 74.2 / 68.3 74.2
- CC-OCR-Bench-V2 79.6
- MTVQA-Test 56.6
- MADQA 91.8
- QwenVisualOffice 44.6
- RealWorldQA 88
- ERQA 77.8
- LingoQA 84.8
- SURDS 77.8
- SimpleVQA 75
- WorldVQA 53.2
- MMStar 85.9
- PerceptionBench 63.5
- CountQA 82.4
- RefAdv-S 80.2
- Dense200 87
- COCO 78.7
- VisFactor 60.8
- VLMsAreBiased 88.3
- VideoMME (w/ Sub.) 90.4
- VideoMME v2 (w/ Sub.) 68.3
- VideoMMMU 88.7
- MMVU 82.4
- MLVU (M-Avg) 90.8
- TVBench 81.9
- LVBench 81.8
- LVBench (w/ Mem.) 85.6
- EgoLife (w/ Mem.) 80.3
- VideoDR (w/ Search) 73.2
Qwen 3.8 Max
Qwen 3.8 Max ranks first on 19 of the 138 benchmarks it has scores on.
170
Benchmarks with a score
19
Ranked first, out of five or more models
38
In the top three
Qwen Aug 2026 Qwen 3.8 Max
Closed weights Flagship
158.0
#22 of 271
Epoch Capabilities Index
- IFBench 82.8% #2 /70
- MMMU-Pro 82.3% #4 /72
- DeepSWE 57.5% #16 /28
Where it ranks
Its best score on each benchmark, and where that ranks among every model with a score there. We rank each model once, at its best setting.
Knowledge and language
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| BullshitBench V2 | 95% | #2 of 122 | Official leaderboard | |
| IFBench | 82.8% | #2 of 70 | Model card | |
| BullshitBench | 87.3% | #6 of 107 | Official leaderboard | |
| LMArena (Elo) | 1,482 | #21 of 333 | Official leaderboard | |
| LongBench v2 | 66.3% | #2 of 27 | Model card | |
| MRCR v2 | 92.9% | #3 of 27 | Model card | |
| Humanity's Last Exam | 56.2% | #19 of 147 | Model card | |
| SimpleVQA | 75% | #3 of 23 | Model card | |
| WideSearch | 81.9% | #3 of 17 | Model card | |
| Extended NYT Connections | 88.3% | #27 of 97 | Official leaderboard | |
| FACTS Grounding v2 | 73.8% | #6 of 15 | Official leaderboard | |
| SimpleQA Verified | 45.8% | #39 of 84 | Epoch AI | |
| Continual Learning Exam | 26.7% | #8 of 10 | Official leaderboard |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OTIS Mock AIME 2024-2025 | 99.4% | #15 of 202 | Epoch AI | |
| GPQA Diamond | 92.7% | #24 of 219 | Epoch AI | |
| xbench-ScienceQA | 73.4% | #5 of 29 | Official leaderboard | |
| ResearchClawBench | 28.4 | #4 of 23 | Official leaderboard | |
| CritPt | 20% | #30 of 160 | Epoch AI | |
| DTBench | 86.7% | #36 of 178 | Official leaderboard | |
| FrontierMath Tiers 1-3 v2 | 74.7% | #22 of 89 | Epoch AI | |
| HiPhO | 90% | #1 of 4 | Model card | |
| ProofBench | 47% | #15 of 59 | Official leaderboard | |
| SimpleBench | 62.5% | #28 of 100 | Official leaderboard | |
| FrontierMath Tier 4 v2 | 46.3% | #22 of 71 | Epoch AI | |
| PhyX | 83.5% | #1 of 3 | Model card | |
| ProofBench v1.1 | 58% | #24 of 51 | Official leaderboard | |
| NatureBench | 11.1% | #9 of 18 | Official leaderboard | |
| ExplorationBench | 65.7% | #6 of 10 | Official leaderboard | |
| MysteryMechanism | 23.9% | #18 of 29 | Official leaderboard | |
| PaperBenchX | 5.38% | #9 of 10 | Official leaderboard | |
| Prophet Arena | 84.3 | #18 of 18 | Official leaderboard |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| IOI (Vals) | 73% | #3 of 55 | Official leaderboard | |
| PaperBench | 93% | #1 of 14 | Model card | |
| WebDev Arena | 1,671 | #10 of 105 | Epoch AI | |
| Code Arena | 1,672 | #14 of 124 | Official leaderboard | |
| SWE-Bench Pro | 67.7% | #10 of 77 | Model card | |
| AndroidBench | 87% | #6 of 32 | Official leaderboard | |
| Image-to-WebDev Arena | 1,615 | #11 of 55 | Official leaderboard | |
| IOI v2 (Vals) | 68.9% | #13 of 45 | Official leaderboard | |
| NL2Repo-Bench | 55.9% | #9 of 28 | Model card | |
| Vision2Web | 69% | #2 of 5 | Model card | |
| MirrorCode | 52.2% | #12 of 28 | Epoch AI | |
| Vibe Code Bench | 64.7% | #51 of 115 | Official leaderboard | |
| SWE Refactor Bench | 10% | #4 of 8 | Official leaderboard | |
| QwenSWEBench | 80.7% | #1 of 2 | Model card | |
| MLS-Bench | 41% | #8 of 15 | Official leaderboard | |
| Vals RSI Index | 0.22 | #14 of 25 | Official leaderboard | |
| DeepSWE | 57.5% | #16 of 28 | Official leaderboard | |
| RSI Bench | 33.4 | #5 of 8 | Official leaderboard | |
| Code Migration | 24% | #49 of 78 | Official leaderboard | |
| Android Bench 2.0 | 12.5% | #7 of 11 | Official leaderboard | |
| FrontierCS 2 | 12.5% | #13 of 16 | Official leaderboard | |
| OpenDesign Arena | 72.0 | #16 of 19 | Official leaderboard | |
| Vibe Code Bench 1-100 | 12.8% | #23 of 26 | Official leaderboard | |
| AutoResearchExam | 0.42 | #8 of 9 | Official leaderboard | |
| FrontierSWE | 15.8% | #19 of 21 | Epoch AI | |
| RSI-Exam | 32.2% | #17 of 18 | Official leaderboard | |
| QwenBlenderBench | 69.9% | #1 of 1 | Model card | |
| QwenQoderBench | 58.4% | #1 of 1 | Model card | |
| QwenReactBench | 1,724 | #1 of 1 | Model card | |
| QwenSVGBench | 1,713 | #1 of 1 | Model card |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| OSWorld-Verified | 86.1% | #1 of 37 | Model card | |
| τ³-bench Banking | 55.1% | #1 of 25 | Official leaderboard | |
| SkillsBench | 70.2% | #1 of 13 | Model card | |
| AndroidWorld | 85.3% | #2 of 18 | Model card | |
| LMArena Search Arena | 1,230 | #5 of 35 | Official leaderboard | |
| ScreenSpot-Pro | 84.5% | #5 of 25 | Model card | |
| Toolathlon | 72.5% | #8 of 37 | Model card | |
| Mystery Game Puzzles | 38% | #18 of 80 | Epoch AI | |
| Terminal-Bench 2.1 | 86.6% | #12 of 53 | Model card | |
| Chess Puzzles | 46% | #24 of 83 | Epoch AI | |
| WebArena-Verified | 66.8% | #2 of 6 | Model card | |
| MobileWorld | 77.8% | #1 of 3 | Model card | |
| RecreationBench | 51.7% | #1 of 3 | Model card | |
| Agent Arena | 0.00 | #30 of 53 | Official leaderboard | |
| Hermes Bench | 70.5% | #11 of 17 | Official leaderboard | |
| AutomationBench | 27.3% | #30 of 41 | Model card | |
| Long-Horizon-Terminal-Bench 2.0 | 43.5% | #12 of 15 | Official leaderboard | |
| OSWorld 2.0 | 19.4% | #18 of 19 | Model card | |
| SaaS-Bench | 23.6% | #4 of 4 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MedXpertQA-MM | 80.4% | #1 of 12 | Model card | |
| TaxEval v2 | 75.5% | #17 of 129 | Official leaderboard | |
| Legal Research Bench | 47.6% | #14 of 88 | Official leaderboard | |
| Immigration Bench | 70.1% | #4 of 24 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 10.4% | #18 of 80 | Official leaderboard | |
| CorpFin v2 | 65.8% | #27 of 119 | Official leaderboard | |
| APEX-Agents | 63.3% | #14 of 56 | Epoch AI | |
| PMC-VQA | 66.2% | #1 of 4 | Model card | |
| SLAKE | 90.8% | #1 of 4 | Model card | |
| E-Commerce Bench | 4.72 | #5 of 19 | Official leaderboard | |
| Tax Agent Bench | 32.1% | #22 of 79 | Official leaderboard | |
| CommerceAgentBench | 52.3% | #4 of 13 | Official leaderboard | |
| JobBench | 52.7% | #5 of 16 | Official leaderboard | |
| Public Benefits Bench | 67.1% | #16 of 50 | Official leaderboard | |
| MedScribe | 85% | #32 of 98 | Official leaderboard | |
| HealthBench | 60.2% | #10 of 30 | Model card | |
| CoWorkBench | 74.8% | #2 of 6 | Model card | |
| Parametric CAD Bench | 91.5% | #1 of 3 | Model card | |
| PresentBench | 79.6% | #1 of 3 | Model card | |
| WorkSpaceBench | 67.7% | #1 of 3 | Model card | |
| LegalBench | 83.6% | #51 of 138 | Official leaderboard | |
| $OneMillion-Bench | 31.9% | #5 of 13 | Official leaderboard | |
| BlenderWorld | 60.6 | #5 of 13 | Official leaderboard | |
| OR-License | 78.7% | #4 of 10 | Official leaderboard | |
| Excel Modeling Benchmark | 60.1% | #35 of 76 | Official leaderboard | |
| Vals Finance Agent v2 | 50.6% | #43 of 80 | Official leaderboard | |
| MortgageTax | 64% | #48 of 86 | Official leaderboard | |
| MedCode | 40.7% | #59 of 96 | Official leaderboard | |
| RWE-bench | 0% | #7 of 11 | Official leaderboard | |
| Epoch Automation Reports | 53% | #4 of 6 | Official leaderboard | |
| Agents' Last Exam | 27% | #17 of 25 | Model card | |
| Excel Modeling Benchmark v1.1 | 66% | #28 of 37 | Official leaderboard | |
| PolicyBench | 74.8% | #45 of 50 | Official leaderboard | |
| PLawBench | 73.2% | #1 of 1 | Model card | |
| PRBench | 58.3% | #1 of 1 | Model card |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Vision Arena | 1,300 | #2 of 136 | Official leaderboard | |
| MATH-Vision (MathVision) | 97.7% | #1 of 42 | Model card | |
| Video-MME | 90.4% | #1 of 42 | Model card | |
| CharXiv | 93.5% | #2 of 67 | Model card | |
| MMStar | 85.9% | #1 of 30 | Model card | |
| Video-MMMU | 88.7% | #1 of 30 | Model card | |
| BabyVision | 91.3% | #1 of 27 | Official leaderboard | |
| MMVU | 82.4% | #1 of 20 | Model card | |
| OCRBench v2 | 74.2% | #1 of 20 | Model card | |
| MMMU-Pro | 82.3% | #4 of 72 | Model card | |
| MLVU | 90.8% | #1 of 16 | Model card | |
| RealWorldQA | 88% | #2 of 30 | Model card | |
| ERQA | 77.8% | #2 of 26 | Model card | |
| LogicVista | 91.9% | #1 of 12 | Model card | |
| ZeroBench | 49% | #3 of 25 | Model card | |
| Claw-Eval-MM | 74.8% | #1 of 7 | Model card | |
| TVBench | 81.9% | #1 of 7 | Model card | |
| LVBench | 81.8% | #4 of 25 | Model card | |
| MTVQA-Test | 56.6% | #1 of 6 | Model card | |
| VideoMME v2 (with subtitles) | 68.3% | #1 of 6 | Model card | |
| VLMsAreBiased | 88.3% | #1 of 5 | Model card | |
| LingoQA | 84.8% | #1 of 4 | Model card | |
| VisFactor | 60.8% | #1 of 4 | Model card | |
| WorldVQA | 53.2% | #3 of 9 | Model card | |
| CountQA | 82.4% | #1 of 3 | Model card | |
| HLE-VL | 52.2% | #1 of 3 | Model card | |
| Humanity's Sixth Sense | 33% | #11 of 29 | Official leaderboard | |
| VoxelBench | 1,785 | #24 of 56 | Official leaderboard | |
| GDP.pdf | 23.2% | #18 of 39 | Epoch AI | |
| Code2Video Benchmark | 1,533 | #9 of 19 | Official leaderboard | |
| Dense 200 (bbox) | 87% | #1 of 2 | Model card | |
| SURDS | 77.8% | #1 of 2 | Model card | |
| PerceptionBench (Moonshot) | 63.5% | #2 of 3 | Model card | |
| Embodied First Exam | 3.95% | #5 of 7 | Official leaderboard | |
| CC-OCR-Bench-V2 | 79.6% | #1 of 1 | Model card | |
| COCO | 78.7% | #1 of 1 | Model card | |
| EgoLife (with memory) | 80.3% | #1 of 1 | Model card | |
| LVBench (with memory) | 85.6% | #1 of 1 | Model card | |
| MADQA | 91.8% | #1 of 1 | Model card | |
| QwenVisualOffice | 44.6% | #1 of 1 | Model card | |
| RefAdv-S | 80.2% | #1 of 1 | Model card | |
| VideoDR (with search) | 73.2% | #1 of 1 | Model card |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| DistressBench | 77.6% | #7 of 42 | Official leaderboard | |
| Last Secure Code Benchmark | 22% | #3 of 6 | Official leaderboard | |
| DeepsecBench | 16.4 | #17 of 28 | Official leaderboard | |
| CyberBench v1.1 (Vals) | 28.6% | #43 of 43 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SuperCLUE | 71.5 | #3 of 18 | Official leaderboard | |
| SAGE (Vals) | 51.3% | #14 of 81 | Official leaderboard | |
| Conceptual Reasoning Index | 59.4 | #35 of 149 | Official leaderboard | |
| LMCA | 46.2% | #35 of 139 | Epoch AI | |
| ACCoRD | 71.4 | #40 of 150 | Official leaderboard | |
| Vals Multimodal Index | 65.4% | #12 of 33 | Official leaderboard | |
| Hermes Index | 36.2 | #12 of 17 | Official leaderboard | |
| LiveBench | 61.6% | #21 of 28 | Official leaderboard | |
| SAGE v1.1 (Vals) | 81.8% | #24 of 28 | Official leaderboard |
Qwen's models over time
Qwen's models in color, against every other model on the Epoch Capabilities Index, by release date.
Its launch post
The benchmarks the lab chose to headline, with the scores it gave.