- Who
- Alibaba
- What
- released the Qwen4-Max foundation model
- When
- Quote
11月16日,阿里巴巴发布全新一代基座大模型Qwen4-Max Qwen4-Max现已上线千问AI平台对外提供API服务,缓存命中每百万Tokens低至0.5元,完成单任务的平均费用较上代模型大降6成 用户也可通过阿里Token Plan调用新模型,或在千问办公、Qoder、千问APP等应用中直接体验。
What happened
Alibaba released Qwen4-Max on November 16, describing it as a 2.4T-parameter model with native vision and a 1M-token context window. In Terminal-Bench 4.0, it scored 48.5 versus 38.9 for Qwen3.8-Max while using an average of 151K tokens per attempt instead of 289K; the report says average cost per task fell 58%. Qwen4-Max is available through the Qianwen AI platform API, and Alibaba says smaller Qwen4 models will be open sourced over time.
Why it matters
Qwen4-Max reduces average task cost by 58% versus Qwen3.8-Max while improving its Terminal-Bench 4.0 score from 38.9 to 48.5, changing the cost and performance tradeoff for professional coding tasks.
Read next
See the full storyAlibaba Releases Qwen4-Max with Native Vision and Live API AccessThe story so far
Alibaba released Qwen4-Max, a 2.4T-parameter MoE model with native vision and a 1M-token context window, with its API now live on QwenCloud and the Qianwen AI platform. The model is also available on Qwen Studio, QwenWork, Qoder, Qwen APP, Alibaba’s Token Plan, OpenRouter, OpenCode Go, Vercel AI Gateway, and Novita Labs. Vercel says its version supports up to 384K output tokens and is priced 25% below Qwen3.8-Max. Qwen lists QwenCloud API pricing at $1.5 per million input tokens, $4.5 per million output tokens, and $0.075 per million tokens for implicit caching, which it says is 70% lower than Qwen3.8-Max. QbitAI reports Qianwen AI platform cache-hit pricing as low as 0.5 yuan per million tokens. Alibaba says open weights for smaller Qwen4 models will be released over time. In Qwen’s evaluations, Qwen4-Max scored 52.69 on Work360, about 10% above Qwen3.8-Max, while cutting token use by about 24% at Xhigh and 35% at High and lowering average task cost from $4.43 to $1.67. On Terminal-Bench 4.0, Qwen says token use fell by about 48% and 49% at those settings, while QbitAI reports a score of 48.5 versus 38.9 for Qwen3.8-Max, with average task cost down 58%.
How the story developed · 2
View full event- 11/16 22:39 (UTC+8)
Alibaba Releases Qwen4-Max with 2.4T Parameters and 1M-Token Context This report
- 11/16 10:38 (UTC+8)Qwen4-Max launches with native vision and a 1M-token context windowQwen · X
