Qwen4 is coming.
Deeper knowledge. Richer world understanding. Stronger coding and agentic capabilities.
Built for real-world tasks — from complex coding and knowledge work to front-end builds and professional software workflows.
Qwen4-Max is now in preview on Alibaba’s Token Plan. You can also get early access to the new model on QwenWork, Qoder, Qwen APP, and QwenStudio.
Put it to the test on a real problem. Show us what you build. 👀
Token Plan: https://t.co/LOB2Jdt3Sk
Word on the street says Qwen4 release has been delayed because the Cyber module hasn’t been fully separated and hasn’t cleared review by relevant Chinese regulators.
Looking at recent CAC rules requiring security filing for open-source model weights, real-name download requirements plus the draft export-control framework, this line of reasoning makes sense.
Already released models face limited impact, but upcoming next-gen open models including DeepSeek, Kimi and GLM will likely encounter similar compliance scrutiny. Their release timelines may get slowed down.