Alibaba Cloud / Qwen Team model product
Qwen3.6 Plus is Alibaba's next-generation flagship model featuring a 1 million token native context window, up to 65,536 output tokens, and always-on chain-of-thought reasoning. It uses a next-generation hybrid architecture optimized for efficiency and scalability. It leads on Terminal-Bench 2.0 agentic coding (61.6), surpassing Claude 4.5 Opus, and achieves strong results on document understanding (OmniDocBench 91.2) and multimodal reasoning (MMMU 86.0). Compared to Qwen 3.5, it is significantly more decisive in reasoning, using fewer tokens on straightforward tasks with better agent stability.
Updated Aug 10, 2026. Default version: Qwen3.6 Plus
Structured fields from the published default version.
This profile uses the latest version under this unique model that has a calculated LLMBoard score. Arena and price are excluded.
Published benchmark records for the scored version Qwen3.6 Plus.
| CC-OCR | 0.8 | 1 | 18 | 100.0% | C | |
| C-Eval | 0.9 | 1 | 18 | 100.0% | C | |
| DynaMath | 0.9 | 1 | 7 | 100.0% | C | |
| MMLongBench-Doc | 0.6 | 1 | 5 | 100.0% | C | |
| MMMU | 0.9 | 1 | 63 | 100.0% | C | |
| MMStar | 0.8 | 1 | 22 | 100.0% | C | |
| ODinW | 0.5 | 1 | 16 | 100.0% | C | |
| RefCOCO-avg | 0.9 | 1 | 7 | 100.0% | C | |
| TIR-Bench | 0.6 | 1 | 4 | 100.0% | C | |
| We-Math | 0.9 | 1 | 1 | 100.0% | C | |
| AI2D | 0.9 | 2 | 32 | 96.8% | C | |
| DeepPlanning | 0.4 | 2 | 9 | 87.5% | C | |
| HMMT25 | 0.9 | 2 | 25 | 95.8% | C | |
| MMLU-Pro | 0.9 | 2 | 129 | 99.2% | C | |
| SuperGPQA | 0.7 | 2 | 34 | 97.0% | C | |
| TAU3-Bench | 0.7 | 2 | 5 | 75.0% | C | |
| V* | 1.0 | 2 | 7 | 83.3% | C | |
| IFEval | 0.9 | 3 | 65 | 96.9% | C | |
| LongBench v2 | 0.6 | 3 | 17 | 87.5% | C | |
| MLVU | 0.9 | 3 | 10 | 77.8% | C | |
| MMLU-Redux | 0.9 | 3 | 48 | 95.7% | C | |
| OmniDocBench 1.5 | 0.9 | 3 | 16 | 86.7% | C | |
| PolyMATH | 0.8 | 3 | 23 | 90.9% | C | |
| AIME 2026 | 1.0 | 4 | 17 | 81.3% | C | |
| CountBench | 1.0 | 4 | 6 | 40.0% | C | |
| Include | 0.9 | 4 | 31 | 90.0% | C | |
| MAXIFE | 0.9 | 4 | 11 | 70.0% | C | |
| MMLU-ProX | 0.8 | 4 | 32 | 90.3% | C | |
| VITA-Bench | 0.4 | 4 | 10 | 66.7% | C | |
| WideSearch | 0.7 | 4 | 9 | 62.5% | C | |
| AA-LCR | 0.7 | 5 | 15 | 71.4% | C | |
| HMMT 2025 | 1.0 | 5 | 33 | 87.5% | C | |
| MCP-Mark | 0.5 | 5 | 8 | 42.9% | C | |
| RealWorldQA | 0.9 | 5 | 26 | 84.0% | C | |
| WMT24++ | 0.8 | 5 | 23 | 81.8% | C | |
| ERQA | 0.7 | 6 | 23 | 77.3% | C | |
| Global PIQA | 0.9 | 6 | 13 | 58.3% | C | |
| MathVision | 0.9 | 6 | 32 | 83.9% | C | |
| NOVA-63 | 0.6 | 6 | 11 | 50.0% | C | |
| SkillsBench | 0.5 | 6 | 7 | 16.7% | C | |
| HMMT Feb 26 | 0.9 | 7 | 11 | 40.0% | C | |
| LiveCodeBench v6 | 0.9 | 7 | 53 | 88.5% | C | |
| SimpleVQA | 0.7 | 7 | 13 | 50.0% | C | |
| Video-MME | 0.8 | 8 | 17 | 56.3% | C | |
| VideoMMMU | 0.8 | 10 | 26 | 64.0% | C | |
| Claw-Eval | 0.6 | 11 | 13 | 16.7% | C | |
| IFBench | 0.7 | 11 | 28 | 63.0% | C | |
| IMO-AnswerBench | 0.8 | 11 | 19 | 44.4% | C | |
| MMMLU | 0.9 | 11 | 49 | 79.2% | C | |
| ScreenSpot Pro | 0.7 | 11 | 24 | 56.5% | C | |
| SWE-bench Multilingual | 0.7 | 11 | 34 | 69.7% | C | |
| NL2Repo | 0.4 | 12 | 14 | 15.4% | C | |
| MCP Atlas | 0.7 | 14 | 30 | 55.2% | C | |
| FrontierSWE | 0.2 | 15 | 15 | 0.0% | B | |
| CharXiv-R | 0.8 | 16 | 47 | 67.4% | C | |
| Finance Agent v2 | 0.4 | 16 | 26 | 40.0% | B | |
| MMMU-Pro | 0.8 | 17 | 65 | 75.0% | C | |
| OSWorld-Verified | 0.6 | 18 | 22 | 19.1% | C | |
| SWE-Bench Verified | 0.8 | 18 | 104 | 83.5% | C | |
| Terminal-Bench 2.0 | 0.6 | 19 | 49 | 62.5% | C | |
| GPQA | 0.9 | 22 | 233 | 91.0% | C | |
| Toolathlon | 0.4 | 25 | 31 | 20.0% | C | |
| SWE-Bench Pro | 0.6 | 26 | 44 | 41.9% | C | |
| LiveBench | 0.7 | 28 | 38 | 27.0% | B | |
| Humanity's Last Exam | 0.3 | 44 | 92 | 52.8% | C |
Preference and agent-evaluation signals from published Arena datasets.
| webdev | image to webdev | 27 | 1470.0 | 5,326 | N/A | |
| text factuality | korean | 28 | 1379.6 | 432 | N/A | |
| text factuality | math | 32 | 1452.5 | 1,995 | N/A | |
| text factuality | japanese | 34 | 1386.5 | 386 | N/A | |
| text | german | 35 | 1453.9 | 755 | N/A | |
| webdev | webdev-react | 39 | 1453.1 | 10,809 | N/A | |
| webdev | overall | 40 | 1458.3 | 14,124 | N/A | |
| webdev | webdev | 40 | 1458.3 | 14,124 | N/A | |
| webdev | webdev-html | 41 | 1462.6 | 1,551 | N/A | |
| text factuality | french | 42 | 1460.3 | 1,383 | N/A | |
| text factuality | industry mathematical | 43 | 1444.7 | 2,053 | N/A | |
| text factuality | spanish | 46 | 1428.2 | 1,121 | N/A | |
| text style control | german | 49 | 1455.2 | 755 | N/A | |
| text | math | 50 | 1450.3 | 2,513 | N/A | |
| text style control | math | 51 | 1454.8 | 2,513 | N/A | |
| text | hard prompts english | 54 | 1455.2 | 14,367 | N/A | |
| text | expert | 55 | 1457.6 | 4,330 | N/A | |
| text | coding | 56 | 1466.3 | 13,660 | N/A | |
| text | industry mathematical | 56 | 1449.3 | 2,587 | N/A | |
| text | industry software and it services | 57 | 1461.6 | 18,768 | N/A | |
| text | longer query | 57 | 1439.8 | 19,118 | N/A | |
| text | russian | 57 | 1435.4 | 4,730 | N/A | |
| text factuality | expert | 57 | 1469.3 | 3,541 | N/A | |
| text style control | industry mathematical | 57 | 1456.5 | 2,587 | N/A | |
| text | hard prompts | 59 | 1448.7 | 30,200 | N/A | |
| text | industry business and management and financial operations | 59 | 1431.2 | 9,401 | N/A | |
| text | instruction following | 59 | 1425.4 | 15,142 | N/A | |
| text style control | expert | 59 | 1478.2 | 4,330 | N/A | |
| text factuality | industry medicine and healthcare | 60 | 1467.7 | 2,493 | N/A | |
| text | industry writing and literature and language | 61 | 1419.2 | 10,371 | N/A | |
| text factuality | industry writing and literature and language | 61 | 1425.9 | 9,471 | N/A | |
| text factuality | industry business and management and financial operations | 62 | 1447.4 | 8,162 | N/A | |
| text | polish | 63 | 1432.1 | 970 | N/A | |
| text style control | hard prompts english | 63 | 1476.4 | 14,367 | N/A | |
| text style control | longer query | 63 | 1454.6 | 19,118 | N/A | |
| text | english | 65 | 1445.8 | 20,397 | N/A | |
| text factuality | russian | 65 | 1441.9 | 3,808 | N/A | |
| text | overall | 66 | 1436.7 | 45,033 | N/A | |
| text | chinese | 66 | 1475.5 | 2,205 | N/A | |
| text | french | 66 | 1450.5 | 1,744 | N/A | |
| text | non english | 66 | 1424.4 | 24,633 | N/A | |
| text factuality | hard prompts english | 66 | 1473.7 | 13,509 | N/A | |
| text style control | coding | 66 | 1495.0 | 13,660 | N/A | |
| text style control | hard prompts | 66 | 1467.5 | 30,200 | N/A | |
| text style control | industry business and management and financial operations | 66 | 1446.1 | 9,401 | N/A | |
| text style control | instruction following | 66 | 1436.7 | 15,142 | N/A | |
| text factuality | chinese | 67 | 1470.2 | 1,774 | N/A | |
| text style control | english | 67 | 1456.1 | 20,397 | N/A | |
| text | exclude ties | 68 | 1430.4 | 33,786 | N/A | |
| text factuality | coding | 68 | 1490.5 | 12,744 | N/A | |
| text factuality | longer query | 69 | 1452.1 | 18,327 | N/A | |
| text factuality | industry entertainment and sports and media | 70 | 1410.5 | 8,514 | N/A | |
| text factuality | instruction following | 70 | 1435.6 | 14,293 | N/A | |
| text style control | french | 70 | 1461.6 | 1,744 | N/A | |
| text style control | industry software and it services | 70 | 1482.4 | 18,768 | N/A | |
| text | japanese | 71 | 1384.7 | 495 | N/A | |
| text | multi turn | 71 | 1440.4 | 7,951 | N/A | |
| text | spanish | 71 | 1432.3 | 1,506 | N/A | |
| text style control | overall | 71 | 1443.3 | 45,033 | N/A | |
| text style control | industry writing and literature and language | 71 | 1423.8 | 10,371 | N/A | |
| text factuality | industry legal and government | 72 | 1448.6 | 2,749 | N/A | |
| text style control | exclude ties | 72 | 1441.7 | 33,786 | N/A | |
| text | industry entertainment and sports and media | 73 | 1402.9 | 9,375 | N/A | |
| text factuality | english | 73 | 1450.8 | 20,331 | N/A | |
| text style control | polish | 73 | 1431.0 | 970 | N/A | |
| text style control | russian | 73 | 1440.1 | 4,730 | N/A | |
| text | creative writing | 74 | 1404.4 | 6,625 | N/A | |
| text factuality | creative writing | 74 | 1411.2 | 5,594 | N/A | |
| text | industry medicine and healthcare | 75 | 1444.7 | 3,159 | N/A | |
| text style control | chinese | 75 | 1480.9 | 2,205 | N/A | |
| text style control | industry entertainment and sports and media | 75 | 1407.2 | 9,375 | N/A | |
| text style control | non english | 75 | 1427.6 | 24,633 | N/A | |
| text | industry legal and government | 76 | 1437.2 | 3,381 | N/A | |
| text | korean | 76 | 1379.8 | 781 | N/A | |
| text style control | japanese | 76 | 1387.1 | 495 | N/A | |
| text style control | creative writing | 77 | 1407.5 | 6,625 | N/A | |
| text style control | industry medicine and healthcare | 77 | 1460.4 | 3,159 | N/A | |
| text factuality | exclude ties | 78 | 1435.9 | 33,718 | N/A | |
| text style control | industry life and physical and social science | 78 | 1455.2 | 7,065 | N/A | |
| text style control | spanish | 78 | 1431.5 | 1,506 | N/A | |
| text factuality | overall | 79 | 1437.4 | 44,947 | N/A | |
| text factuality | industry software and it services | 79 | 1474.1 | 18,191 | N/A | |
| text style control | multi turn | 79 | 1447.3 | 7,951 | N/A | |
| text factuality | hard prompts | 80 | 1459.9 | 30,140 | N/A | |
| text factuality | non english | 80 | 1421.5 | 24,580 | N/A | |
| text style control | industry legal and government | 80 | 1446.7 | 3,381 | N/A | |
| text | industry life and physical and social science | 82 | 1441.8 | 7,065 | N/A | |
| text factuality | industry life and physical and social science | 82 | 1456.0 | 6,008 | N/A | |
| text style control | korean | 82 | 1381.7 | 781 | N/A | |
| text factuality | multi turn | 86 | 1437.0 | 6,730 | N/A |
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| Alibaba Coding Plan (China) | qwen3.6-plus | global | N/A | N/A | 1M | |
| Alibaba Coding Plan | qwen3.6-plus | global | N/A | N/A | 1M | |
| Alibaba Token Plan | qwen3.6-plus | global | N/A | N/A | 1M | |
| Alibaba Token Plan (China) | qwen3.6-plus | global | N/A | N/A | 1M | |
| AIHubMix | qwen3.6-plus | global | $0.28 | $1.69 | 991K | |
| Pioneer | qwen3.6-plus | global | $0.325 | $1.95 | 1M | |
| OpenCode Go | qwen3.6-plus | global | $0.50 | $3 | 1M | |
| LLM Gateway | qwen3.6-plus | global | $0.50 | $3 | 262.1K | |
| OpenCode Zen | qwen3.6-plus | global | $0.50 | $3 | 262.1K | |
| Alibaba (China) | qwen3.6-plus | global | $0.50 | $3 | 1M | |
| Charm Hyper | qwen3.6-plus | global | $2 | $6 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
All published versions linked to this unique model. The score columns identify the version used by the current overall ranking.
| Qwen3.6 Plus | 72.4 | N/A | 1M | 65.5K | No | Proprietary |
A concise description based on the published model registry.
Qwen3.6 Plus is Alibaba's next-generation flagship model featuring a 1 million token native context window, up to 65,536 output tokens, and always-on chain-of-thought reasoning. It uses a next-generation hybrid architecture optimized for efficiency and scalability. It leads on Terminal-Bench 2.0 agentic coding (61.6), surpassing Claude 4.5 Opus, and achieves strong results on document understanding (OmniDocBench 91.2) and multimodal reasoning (MMMU 86.0). Compared to Qwen 3.5, it is significantly more decisive in reasoning, using fewer tokens on straightforward tasks with better agent stability.
Use the benchmark, Arena and pricing sections above as separate evidence. A missing field means the current data snapshot does not support that claim.
Data snapshot: 2026-08-07. Editorial model content is not available in the backend.
Open a comparison with the three ranked models immediately above and below this model.
Recommendations prioritize the same model type and family, then the closest published LLMBoard score.
Common questions about Qwen3.6 Plus.
Qwen3.6 Plus's default version was released on Mar 31, 2026.
Qwen3.6 Plus's official API price is $0.50 per million input tokens and $3 per million output tokens via Alibaba (China). The lowest tracked third-party offer starts at $0.28 input and $1.69 output via AIHubMix.
Qwen3.6 Plus is published under Alibaba Cloud / Qwen Team in the model registry.
The default version has a 1M token context window.
No. The default version is not marked as having publicly available weights.
11 published provider offerings are linked to the default version.
Nearby ranked alternatives include Claude Sonnet 3.5, Qwen3.7 Plus, Gemini 3 Flash.