Alibaba Cloud / Qwen Team model product
Qwen3.7 Max is Alibaba Cloud Qwen Team's proprietary flagship model for agent-driven workflows. It is designed for coding agents, office automation, MCP and multi-agent orchestration, and long-horizon autonomous execution, with a 1 million token context window and up to 65,536 output tokens. Qwen reports strong agentic coding results including 69.7 on Terminal-Bench 2.0-Terminus, 80.4 on SWE-bench Verified, 60.6 on SWE-Pro, and 78.3 on SWE-Multilingual, alongside 92.4 on GPQA Diamond and 97.1 on HMMT 2026 Feb.
Updated Aug 10, 2026. Default version: Qwen3.7 Max
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 Qwen3.7 Max.
| BFCL-V4 | 0.8 | 1 | 13 | 100.0% | C | |
| HMMT Feb 26 | 1.0 | 1 | 11 | 100.0% | C | |
| Kernel Bench L3 | 1.0 | 1 | 1 | 100.0% | C | |
| LiveCodeBench v6 | 0.9 | 1 | 53 | 100.0% | C | |
| MAXIFE | 0.9 | 1 | 11 | 100.0% | C | |
| MMLU-Pro | 0.9 | 1 | 129 | 100.0% | C | |
| MMLU-ProX | 0.9 | 1 | 32 | 100.0% | C | |
| MMLU-Redux | 0.9 | 1 | 48 | 100.0% | C | |
| MRCR 128K (8-needle) | 0.9 | 1 | 2 | 100.0% | C | |
| PolyMATH | 0.9 | 1 | 23 | 100.0% | C | |
| QwenSVG | 1608.0 | 1 | 1 | 100.0% | C | |
| QwenWebBench | 1568.0 | 1 | 2 | 100.0% | C | |
| SuperGPQA | 0.7 | 1 | 34 | 100.0% | C | |
| ZClawBench | 0.6 | 1 | 4 | 100.0% | C | |
| CoWorkBench | 0.7 | 2 | 3 | 50.0% | C | |
| CritPT | 0.1 | 2 | 4 | 66.7% | C | |
| Include | 0.9 | 2 | 31 | 96.7% | C | |
| MCP-Mark | 0.6 | 2 | 8 | 85.7% | C | |
| NOVA-63 | 0.6 | 2 | 11 | 90.0% | C | |
| QwenWorldBench | 0.6 | 2 | 2 | 0.0% | C | |
| SkillsBench | 0.6 | 2 | 7 | 83.3% | C | |
| SpreadSheetBench-v1 | 0.9 | 2 | 3 | 50.0% | C | |
| VITA-Bench | 0.5 | 2 | 10 | 88.9% | C | |
| Global PIQA | 0.9 | 3 | 13 | 83.3% | C | |
| MathArena Apex | 0.4 | 3 | 7 | 66.7% | C | |
| WMT24++ | 0.9 | 3 | 23 | 90.9% | C | |
| IFEval | 0.9 | 4 | 65 | 95.3% | C | |
| IMO-AnswerBench | 0.9 | 4 | 19 | 83.3% | C | |
| NL2Repo | 0.5 | 4 | 14 | 76.9% | C | |
| SciCode | 0.5 | 4 | 18 | 82.3% | C | |
| SWE-bench Multilingual | 0.8 | 4 | 34 | 90.9% | C | |
| Claw-Eval | 0.7 | 5 | 13 | 66.7% | C | |
| IFBench | 0.8 | 5 | 28 | 85.2% | C | |
| Finance Agent v2 | 0.5 | 8 | 26 | 72.0% | B | |
| MMMLU | 0.9 | 8 | 49 | 85.4% | C | |
| Terminal-Bench 2.0 | 0.7 | 9 | 49 | 83.3% | C | |
| MCP Atlas | 0.8 | 10 | 30 | 69.0% | C | |
| SWE-Bench Verified | 0.8 | 11 | 104 | 90.3% | C | |
| SWE-Bench Pro | 0.6 | 13 | 44 | 72.1% | C | |
| GPQA | 0.9 | 14 | 233 | 94.4% | C | |
| LiveBench | 0.7 | 19 | 38 | 51.4% | B | |
| Humanity's Last Exam | 0.4 | 31 | 92 | 67.0% | C |
Preference and agent-evaluation signals from published Arena datasets.
The default version has no published Arena rows, or its source alias has not been resolved.
Official vendor API PAYG pricing is summarized first. The table then lists individual provider offerings without treating their minimum as the official price.
| Alibaba Token Plan | qwen3.7-max | global | N/A | N/A | 1M | |
| Alibaba Token Plan (China) | qwen3.7-max | global | N/A | N/A | 1M | |
| LLM Gateway | qwen3.7-max | global | $1.25 | $3.75 | 1M | |
| Pioneer | qwen3.7-max | global | $1.25 | $3.75 | 1M | |
| AIHubMix | qwen3.7-max | global | $1.69 | $5.07 | 991K | |
| OpenCode Go | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| NanoGPT | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Alibaba Coding Plan (China) | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Alibaba Coding Plan | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Charm Hyper | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Alibaba (China) | qwen3.7-max | global | $2.5 | $7.5 | 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.
| Qwen3.7 Max | 90.3 | N/A | 1M | 65.5K | No | Proprietary |
A concise description based on the published model registry.
Qwen3.7 Max is Alibaba Cloud Qwen Team's proprietary flagship model for agent-driven workflows. It is designed for coding agents, office automation, MCP and multi-agent orchestration, and long-horizon autonomous execution, with a 1 million token context window and up to 65,536 output tokens. Qwen reports strong agentic coding results including 69.7 on Terminal-Bench 2.0-Terminus, 80.4 on SWE-bench Verified, 60.6 on SWE-Pro, and 78.3 on SWE-Multilingual, alongside 92.4 on GPQA Diamond and 97.1 on HMMT 2026 Feb.
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 Qwen3.7 Max.
Qwen3.7 Max's default version was released on May 19, 2026.
Qwen3.7 Max's official API price is $2.5 per million input tokens and $7.5 per million output tokens via Alibaba (China). The lowest tracked third-party offer starts at $1.25 input and $3.75 output via LLM Gateway.
Qwen3.7 Max 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 GPT-5.6-Sol, Seed 2.1 Pro, GPT-5.6-Terra.