Loading…
Worth opening first
Standouts
Built something with Qwen4-Max Preview?
Paste the URL and Dropday.ai will credit you, track the build as it evolves, and put it on this drop.
Submit a buildGet the best builds by email. No account needed.
Live drop
Alibaba (Qwen)Dropped Nov 11, 20261 build1 maker
1M-context multimodal input with native thinking for extended agentic coding and video/document analysis.
About this release
Qwen4-Max-Preview is Alibaba's first public preview model from the Qwen4 family. It accepts text, images and video inputs, features a 1M-token context window and up to 393k output tokens (shared between thinking and final answer), and defaults to on-by-default reasoning with adjustable effort levels. It supports tool calling, structured outputs and OpenAI/Anthropic API compatibility, aimed at coding, frontend work and agent tasks. The preview runs only through Token Plan access until November 25, 2026, after which it retires.
Read the official Alibaba (Qwen) launch pageWorth opening first
Paste the URL and Dropday.ai will credit you, track the build as it evolves, and put it on this drop.
Submit a build