Qwen4.0-Max is in development, based on a GitHub pull request referencing it
2 sources, first seen
See it on the rumor trackerRumor
Reddit post claims Qwen4 models will have a 1 million token context window
A community post says Qwen4 models will support a 1M context window and points to a GitHub pull request, with a Qwen4-Max date that the poster says is likely a placeholder.
Every outlet that ran it 1 article
r/LocalLLaMACommunityLead