- SWE-bench Verified 78.8
- SWE-bench Multilingual 73.8
- SWE-bench Pro 56.6
- Terminal-Bench 2.0 61.6
- Claw-Eval Avg 74.8
- Claw-Eval Pass^3 58.7
- SkillsBench Avg5 45.7
- QwenClawBench 57.2
- NL2Repo 37.9
- QwenWebBench 1501.7
- TAU3-Bench 70.7
- VITA-Bench 44.3
- DeepPlanning 41.5
- Tool Decathlon 39.8
- MCPMark 48.2
- MCP-Atlas 74.1
- Best of HLE w/ tool / HLE 50.6
- WideSearch 74.3
- MMLU-Pro 88.5
- MMLU-Redux 94.5
- SuperGPQA 71.6
- C-Eval 93.3
- IFEvalstrict prompt 94.3
- IFBench 74.2
- AA-LCR 68.3
- LongBench v2 62
- GPQA 90.4
- LiveCodeBench v6 87.1
- HMMT Feb 25 96.7
- HMMT Nov 25 94.6
- HMMT Feb 26 87.8
- IMOAnswerBench 83.8
- AIME26 95.3
- MMMLU 89.5
- MMLU-ProX 84.7
- NOVA-63 57.9
- INCLUDE 85.1
- Global PIQA 89.8
- PolyMATH 77.4
- WMT24++ 84.3
- MAXIFE 88.2
- MMMU 86
- MMMU-Pro 78.8
- MathVision 88
- We-Math 89
- DynaMath 88
- RealWorldQA 85.4
- MMStar 83.3
- SimpleVQA 67.3
- OmniDocBench1.5 91.2
- CharXiv(RQ) 81.5
- MMLongBench-Doc 62
- CC-OCR 83.4
- AI2D_TEST 94.4
- CountBench 97.6
- ODinW13 51.8
- ERQA 65.7
- V* (96.9 / 90.5) 96.9
- Best of VideoMME(w sub.) / VideoMME(w/o sub.) 87.8
- VideoMMMU 84
- MLVU (M-Avg) 86.7
- ScreenSpot Pro 68.2
- TIR-Bench (61.6 / 43.7) 61.6
- OSWorld-Verified 62.5
Qwen 3.6 Plus
Qwen 3.6 Plus ranks first on 5 of the 117 benchmarks it has scores on.
120
Benchmarks with a score
5
Ranked first, out of five or more models
16
In the top three
Qwen Mar 2026 Qwen 3.6 Plus
Closed weights Flagship
147.8
#72 of 271
Epoch Capabilities Index
- SWE-bench Verified 57.9% #29 /32
- IFBench 74.2% #25 /70
- MMMU-Pro 78.8% #17 /72
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 | |
|---|---|---|---|---|
| MMLU-Pro | 88.5% | #3 of 151 | Model card | |
| C-Eval | 93.3% | #2 of 38 | Model card | |
| IFEval | 94.3% | #7 of 126 | Model card | |
| MMLU-Redux | 94.5% | #3 of 38 | Model card | |
| SuperGPQA | 71.6% | #3 of 26 | Model card | |
| BullshitBench V2 | 72% | #17 of 122 | Official leaderboard | |
| MMMLU | 89.5% | #11 of 62 | Model card | |
| LMArena (Elo) | 1,443 | #70 of 333 | Official leaderboard | |
| Humanity's Last Exam | 50.6% | #34 of 147 | Model card | |
| MMLU-ProX | 84.7% | #4 of 17 | Model card | |
| LongBench v2 | 62% | #7 of 27 | Model card | |
| INCLUDE | 85.1% | #7 of 25 | Model card | |
| CL-bench | 20.3% | #8 of 24 | Official leaderboard | |
| AA-LCR (Artificial Analysis Long Context Reasoning) | 68.3% | #11 of 31 | Model card | |
| IFBench | 74.2% | #25 of 70 | Model card | |
| WideSearch | 74.3% | #8 of 17 | Model card | |
| SimpleVQA | 67.3% | #11 of 23 | Model card | |
| SimpleQA Verified | 44.1% | #42 of 84 | Epoch AI | |
| WMT24++ | 84.3% | #7 of 12 | Model card | |
| Extended NYT Connections | 60.3% | #60 of 97 | Official leaderboard | |
| MAXIFE | 88.2% | #4 of 5 | Model card | |
| Global PIQA | 89.8% | #7 of 8 | Model card | |
| NOVA-63 | 57.9% | #5 of 5 | Model card |
Math and science
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| PolyMath | 77.4% | #3 of 18 | Model card | |
| HMMT Nov 2025 | 94.6% | #4 of 23 | Model card | |
| HMMT 2025 | 96.7% | #11 of 49 | Model card | |
| GPQA Diamond | 88.4% | #53 of 219 | Epoch AI | |
| OTIS Mock AIME 2024-2025 | 93.3% | #49 of 202 | Epoch AI | |
| AIME 2026 | 95.3% | #10 of 34 | Model card | |
| FrontierMath | 26.2% | #23 of 70 | Epoch AI | |
| FrontierMath Tier 4 | 8.33% | #21 of 55 | Epoch AI | |
| DTBench | 69.8% | #81 of 178 | Official leaderboard | |
| IMO-AnswerBench | 83.8% | #17 of 37 | Model card | |
| CritPt | 2.86% | #81 of 160 | Epoch AI | |
| ResearchClawBench | 18.0 | #13 of 23 | Official leaderboard | |
| HMMT 2026 | 87.8% | #11 of 18 | Model card | |
| FrontierMath Tiers 1-3 v2 | 38.3% | #61 of 89 | Epoch AI |
Coding
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| LiveCodeBench v6 | 87.1% | #15 of 76 | Model card | |
| SWE-bench Multilingual | 73.8% | #31 of 70 | Model card | |
| OpenHands Index | 52.9% | #16 of 33 | Official leaderboard | |
| QwenWebBench | 1,502 | #2 of 4 | Model card | |
| WebDev Arena | 1,461 | #53 of 105 | Epoch AI | |
| SWE-Bench Pro | 56.6% | #39 of 77 | Model card | |
| Code Arena | 1,462 | #64 of 124 | Official leaderboard | |
| EvoCode-Bench | 29.4% | #7 of 13 | Official leaderboard | |
| Vibe Code Bench | 25.6% | #76 of 115 | Official leaderboard | |
| EvoCode-Bench v2 | 9.70% | #9 of 13 | Official leaderboard | |
| ALE-Bench | 670 | #85 of 121 | Epoch AI | |
| Image-to-WebDev Arena | 1,472 | #40 of 55 | Official leaderboard | |
| Code Migration | 11.1% | #64 of 78 | Official leaderboard | |
| NL2Repo-Bench | 37.9% | #23 of 28 | Model card | |
| SWE-bench Verified | 57.9% | #29 of 32 | Epoch AI |
Agents and tools
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| DeepTerminalBench | 32% | #2 of 9 | Official leaderboard | |
| SnakeBench | 28 | #48 of 183 | Official leaderboard | |
| Terminal-Bench 2.0 | 61.6% | #18 of 61 | Model card | |
| Pencil Puzzle Bench | 10% | #17 of 53 | Official leaderboard | |
| tau3-bench | 70.7% | #4 of 12 | Model card | |
| DeepPlanning | 41.5% | #2 of 6 | Model card | |
| LiveClawBench | 70.6% | #6 of 17 | Official leaderboard | |
| Toolathlon | 39.8% | #19 of 50 | Official leaderboard | |
| VITA-Bench | 44.3% | #5 of 12 | Model card | |
| Claw-Eval Pass^3 | 58.8% | #11 of 23 | Official leaderboard | |
| PinchBench 2.0 | 72.5% | #29 of 57 | Official leaderboard | |
| MCP Atlas | 74.1% | #22 of 40 | Model card | |
| ScreenSpot-Pro | 68.2% | #14 of 25 | Model card | |
| Claw-Eval | 75% | #13 of 23 | Official leaderboard | |
| Long-Horizon-Terminal-Bench (LHTB) | 31.3% | #16 of 25 | Official leaderboard | |
| QwenClawBench | 57.2% | #4 of 6 | Model card | |
| MCPMark | 48.2% | #7 of 10 | Model card | |
| PinchBench | 63.3% | #35 of 48 | Official leaderboard | |
| MobileGym | 45.7% | #3 of 4 | Official leaderboard | |
| SkillsBench | 45.7% | #10 of 13 | Model card | |
| OSWorld-Verified | 62.5% | #29 of 37 | Model card | |
| Mystery Game Puzzles | 12% | #64 of 80 | Epoch AI | |
| Chess Puzzles | 14% | #70 of 83 | Epoch AI | |
| Claw-Eval-Live | 50.5% | #11 of 12 | Official leaderboard |
Professional work
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| TaxEval v2 | 74.7% | #31 of 129 | Official leaderboard | |
| MortgageTax | 68% | #21 of 86 | Official leaderboard | |
| LegalBench | 84.2% | #37 of 138 | Official leaderboard | |
| Vals Finance Agent | 54.6% | #15 of 49 | Official leaderboard | |
| CorpFin v2 | 61.9% | #51 of 119 | Official leaderboard | |
| Vending-Bench 2 | $5,115 | #33 of 66 | Official leaderboard | |
| Clipping Bench | $458 | #12 of 19 | Official leaderboard | |
| MedScribe | 77% | #66 of 98 | Official leaderboard | |
| Harvey LAB (Legal Agent Benchmark) | 1.25% | #55 of 80 | Official leaderboard | |
| FORTE | 62.5% | #15 of 21 | Official leaderboard | |
| MedCode | 36.9% | #71 of 96 | Official leaderboard | |
| PhysicianBench | 13.7% | #10 of 13 | Official leaderboard | |
| Vals Finance Agent v2 | 40.9% | #63 of 80 | Official leaderboard | |
| Excel Modeling Benchmark | 32.9% | #63 of 76 | Official leaderboard | |
| Legal Research Bench | 14.9% | #73 of 88 | Official leaderboard | |
| Tax Agent Bench | 7.68% | #69 of 79 | Official leaderboard | |
| CaseLaw v2 | 51.5% | #49 of 53 | Official leaderboard | |
| E-Commerce Bench | 0.47 | #18 of 19 | Official leaderboard |
Vision, audio and robotics
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| MMMU | 86% | #1 of 97 | Model card | |
| DynaMath | 88% | #1 of 17 | Model card | |
| V* | 96.9% | #1 of 12 | Model card | |
| We-Math | 89% | #1 of 11 | Model card | |
| MMStar | 83.3% | #3 of 30 | Model card | |
| CountBench | 97.6% | #2 of 20 | Model card | |
| ODinW-13 | 51.8% | #1 of 9 | Model card | |
| AI2D | 94.4% | #6 of 44 | Model card | |
| MMLongBench-Doc | 62% | #2 of 13 | Model card | |
| Video-MME | 87.8% | #8 of 42 | Model card | |
| MMMU-Pro | 78.8% | #17 of 72 | Model card | |
| CC-OCR | 83.4% | #2 of 8 | Model card | |
| MATH-Vision (MathVision) | 88% | #11 of 42 | Model card | |
| RealWorldQA | 85.4% | #8 of 30 | Model card | |
| CharXiv | 81.5% | #24 of 67 | Model card | |
| MLVU | 86.7% | #6 of 16 | Model card | |
| ERQA | 65.7% | #11 of 26 | Model card | |
| Video-MMMU | 84% | #13 of 30 | Model card | |
| TIR-Bench | 61.6% | #1 of 2 | Model card | |
| VoxelBench | 1,281 | #51 of 56 | Official leaderboard |
Safety and security
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| Dictatorship Eval | 38.8% | #15 of 20 | Official leaderboard |
Other
| Benchmark | Score | Rank | Source | |
|---|---|---|---|---|
| SAGE (Vals) | 44.9% | #42 of 81 | Official leaderboard | |
| LMCA | 33.1% | #76 of 139 | Epoch AI | |
| Conceptual Reasoning Index | 45.7 | #84 of 149 | Official leaderboard | |
| ACCoRD | 59.3 | #90 of 150 | Official leaderboard | |
| Vals Multimodal Index | 51.5% | #25 of 33 | 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.