Vercel has added Alibaba’s Qwen4-Max to AI Gateway. Vercel says the reasoning model is the first in the Qwen4 series and has native vision and a 1M-token context window. It is available for coding, professional work and long-running agent tasks.
Vercel says Qwen4-Max can use tools in Blender, Unity, Unreal Engine and Godot to build 3D scenes or playable games. The model can run what it builds, then use visual feedback to address camera, animation and lighting problems.
For finance, legal and healthcare tasks, Vercel says the model can call multiple MCP tools, work through large collections of documents and images, and produce spreadsheets, slide decks and reports with source citations.
Read nextGoogle adds Wallet spending insights to GeminiThe model refines long jobs in iterations
For longer jobs, Vercel says Qwen4-Max produces a usable first version early, then improves it in later iterations. That gives users a chance to review progress and change direction while work continues. Alibaba is open-sourcing a Live Delivery plugin for this workflow.
Vercel also says Qwen4-Max uses fewer tokens than earlier Qwen Max models to complete the same tasks. According to the company, it writes more secure code by default, with fewer injection, memory safety and access control vulnerabilities.

Developers can access the model using Vercel CLI
Developers can select alibaba/qwen4-max as the model in AI Gateway. To use it with coding agents including Claude Code, Codex and Cursor, Vercel says to install the latest Vercel CLI, run ai-gateway setup and select the model in the agent. Qwen4-Max is also available in the model playground.
AI Gateway offers a unified API for model calls, usage and cost tracking, and routing and reliability settings. Vercel says it passes through provider pricing without a markup or platform fee on inference, including requests made with a user’s own key. Qwen4-Max is available on AI Gateway now.



