Alibaba Cloud / Qwen Team model product
Qwen3.7 Max
7 Max is Alibaba Cloud Qwen Team's proprietary flagship model for agent-driven workflows.
Updated Aug 12, 2026. Default version: Qwen3.7 Max
LLMBoard score78.8Qwen3.7 Max
Coverage100%37 benchmark families
Context window1MTokens
Official input price$2.5Alibaba API
Qwen3.7 Max Specifications
Technical details for the model's default version.
- Version
- Qwen3.7 Max
- Released
- May 19, 2026
- Knowledge cutoff
- Unknown
- Parameters
- N/A
- Context window
- 1M
- Max output
- 65.5K
- Inputs
- text
- Outputs
- text
- Open weights
- No
- License
- Proprietary
Qwen3.7 Max Capability Profile
This profile uses the model's current scored version. Arena ratings and prices are shown separately.
Qwen3.7 Max Benchmark Results
Benchmark scores for Qwen3.7 Max.
| BFCL-V4 | 75.0% | 01 | 13 | 100.0% | C | |
| HMMT Feb 26 | 97.1% | 01 | 11 | 100.0% | C | |
| Kernel Bench L3 | 96.0% | 01 | 1 | 100.0% | C | |
| LiveCodeBench v6 | 91.6% | 01 | 53 | 100.0% | C | |
| MAXIFE | 89.2% | 01 | 11 | 100.0% | C | |
| MMLU-Pro | 89.6% | 01 | 129 | 100.0% | C | |
| MMLU-ProX | 87.0% | 01 | 32 | 100.0% | C | |
| MMLU-Redux | 95.0% | 01 | 48 | 100.0% | C | |
| MRCR 128K (8-needle) | 90.4% | 01 | 2 | 100.0% | C | |
| PolyMATH | 86.5% | 01 | 23 | 100.0% | C | |
| QwenSVG | 1,608 points | 01 | 1 | 100.0% | C | |
| QwenWebBench | 1,568 points | 01 | 2 | 100.0% | C | |
| SuperGPQA | 73.6% | 01 | 34 | 100.0% | C | |
| ZClawBench | 64.3% | 01 | 4 | 100.0% | C | |
| CoWorkBench | 67.2% | 02 | 3 | 50.0% | C | |
| CritPT | 11.4% | 02 | 4 | 66.7% | C | |
| Include | 86.2% | 02 | 31 | 96.7% | C | |
| MCP-Mark | 60.8% | 02 | 8 | 85.7% | C | |
| NOVA-63 | 59.0% | 02 | 11 | 90.0% | C | |
| QwenWorldBench | 57.3% | 02 | 2 | 0.0% | C | |
| SkillsBench | 59.2% | 02 | 8 | 85.7% | C | |
| SpreadSheetBench-v1 | 87.0% | 02 | 3 | 50.0% | C | |
| VITA-Bench | 47.9% | 02 | 10 | 88.9% | C | |
| Global PIQA | 91.4% | 03 | 13 | 83.3% | C | |
| MathArena Apex | 44.5% | 03 | 7 | 66.7% | C | |
| WMT24++ | 85.8% | 03 | 23 | 90.9% | C | |
| IFEval | 94.3% | 04 | 65 | 95.3% | C | |
| IMO-AnswerBench | 90.0% | 04 | 19 | 83.3% | C | |
| NL2Repo | 47.2% | 04 | 14 | 76.9% | C | |
| SciCode | 53.5% | 04 | 19 | 83.3% | C | |
| SWE-bench Multilingual | 78.3% | 04 | 34 | 90.9% | C | |
| Claw-Eval | 65.2% | 05 | 13 | 66.7% | C | |
| IFBench | 79.1% | 05 | 29 | 85.7% | C | |
| Finance Agent v2 | 48.4% | 08 | 26 | 72.0% | B | |
| MMMLU | 90.3% | 08 | 49 | 85.4% | C | |
| Terminal-Bench 2.0 | 69.7% | 09 | 49 | 83.3% | C | |
| MCP Atlas | 76.4% | 10 | 31 | 70.0% | C | |
| SWE-Bench Verified | 80.4% | 11 | 105 | 90.4% | C | |
| SWE-Bench Pro | 60.6% | 13 | 45 | 72.7% | C | |
| GPQA | 92.4% | 14 | 234 | 94.4% | C | |
| LiveBench | 74.3% | 19 | 38 | 51.4% | B | |
| Humanity's Last Exam | 41.4% | 31 | 93 | 67.4% | C |
Qwen3.7 Max Arena Results
Preference and agent-evaluation results for the default version.
| text factuality | overall | 13 | 1479.2 | 3,719 | N/A | |
| text | overall | 17 | 1474.5 | 3,719 | N/A | |
| agent | overall | 25 | -0.0 | N/A | 1.1M | |
| agent bash recovery steps | overall | 25 | 0.0 | N/A | 25.3K | |
| agent steerability | overall | 25 | -0.0 | N/A | 31.1K | |
| text style control | overall | 25 | 1473.9 | 3,719 | N/A | |
| webdev | overall | 25 | 1517.3 | 8,368 | N/A | |
| agent task outcome explicit | overall | 28 | -0.0 | N/A | 19K | |
| agent tool hallucination | overall | 30 | 0.0 | N/A | 1M | |
| agent praise complaint | overall | 31 | -0.1 | N/A | 7.3K |
Qwen3.7 Max Runtime Performance
Provider-specific output speed and catalog latency for Qwen3.7 Max. Runtime does not affect the capability score.
| Novita | 10.109 tok/s | 1.946 s | 1M | 65.5K | |
| Together | 5.637 tok/s | 4.203 s | 1M | 65.5K |
Output Speed is generated output tokens received per second. Catalog latency is reported separately from observed provider TTFT.
Qwen3.7 Max Pricing
Official vendor API pricing appears first, followed by individual provider offers.
- Official API
- $2.5 input, $7.5 output per 1M
- Official provider
- Alibaba
- Lowest third-party
- From $0.825 input, $2.48 output per 1M via Merge Gateway
- Tracked offerings
- 28
| 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 | |
| Merge Gateway | qwen/qwen3.7-max | global | $0.825 | $2.48 | 1M | |
| LLM Gateway | qwen3.7-max | global | $1.25 | $3.75 | 1M | |
| Pioneer | qwen3.7-max | global | $1.25 | $3.75 | 1M | |
| Together AI | Qwen/Qwen3.7-Max | global | $1.25 | $3.75 | 1M | |
| Kilo Gateway | qwen/qwen3.7-max | global | $1.25 | $3.75 | 1M | |
| NovitaAI | qwen/qwen3.7-max | global | $1.25 | $3.75 | 1M | |
| OpenRouter | qwen/qwen3.7-max | global | $1.48 | $4.43 | 1M | |
| CrossModel | qwen/qwen3.7-max | global | $1.5 | $4.5 | 1M | |
| AIHubMix | qwen3.7-max | global | $1.69 | $5.07 | 991K | |
| Impossibl | qwen/qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| 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 | |
| ClinePass | cline-pass/qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| ZenMux | qwen/qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| EmpirioLabs AI | qwen3-7-max | global | $2.5 | $7.5 | 1M | |
| Ofox | bailian/qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| GMI Cloud | Qwen/Qwen3.7-Max | global | $2.5 | $7.5 | 1M | |
| Charm Hyper | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Alibaba | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Vercel AI Gateway | alibaba/qwen3.7-max | global | $2.5 | $7.5 | 991K | |
| Alibaba (China) | qwen3.7-max | global | $2.5 | $7.5 | 1M | |
| Deep Infra | Qwen/Qwen3.7-Max | global | $2.5 | $7.5 | 256K | |
| Venice AI | qwen-3-7-max | global | $2.7 | $8.05 | 1M | |
| Modelis | qwen/qwen3.7-max | global | $3 | $9 | 1M |
Official prices use only the vendor's configured official Provider and positive standard USD PAYG rates. Third-party offers remain explicitly labeled.
Qwen3.7 Max Versions
Available versions of this model. The score column identifies the version used in the overall ranking.
| Qwen3.7 Max | 78.8 | N/A | 1M | 65.5K | No | Proprietary |
What is Qwen3.7 Max?
Key information about Qwen3.7 Max and its available data.
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. 1 on HMMT 2026 Feb.
Data as of 2026-08-11.
Qwen3.7 Max vs nearby models
Open a comparison with the three ranked models immediately above and below this model.
Models similar to Qwen3.7 Max
Recommendations prioritize the same model type and family, then the closest LLMBoard score.
AC
Qwen3.7 Plus
Alibaba Cloud / Qwen Team
AC
Qwen3.8 Max
Alibaba Cloud / Qwen Team
AC
Qwen3.6 Plus
Alibaba Cloud / Qwen Team
AC
Qwen3.5 397B A17B
Alibaba Cloud / Qwen Team
AC
Qwen3.6 27B
Alibaba Cloud / Qwen Team
AC
Qwen3.5 122B A10B
Alibaba Cloud / Qwen Team
FAQ
Common questions about Qwen3.7 Max.
When was Qwen3.7 Max released?
Qwen3.7 Max's default version was released on May 19, 2026.
How much does Qwen3.7 Max cost?
Qwen3.7 Max's official API price is $2.5 per million input tokens and $7.5 per million output tokens via Alibaba. The lowest tracked third-party offer starts at $0.825 input and $2.48 output via Merge Gateway.
Who created Qwen3.7 Max?
Qwen3.7 Max was created by Alibaba Cloud / Qwen Team.
What is the context window for Qwen3.7 Max?
The default version has a 1M token context window.
Is Qwen3.7 Max open weight?
No. The default version is not marked as having publicly available weights.
How many API providers offer Qwen3.7 Max?
28 provider offerings are linked to the default version.
What models should I compare Qwen3.7 Max with?
Nearby ranked alternatives include Gemini 3.6 Flash, GLM 5.2, Gemini 3.5 Flash.