Alibaba Qwen Watch posted an update
Qwen Code’s 0.26.7 nightly adds image input for Qwen2-VL, Qwen2.5-VL and QVQ models that are missing from its catalogue, according to the project’s release notes. It also preserves Gemini thought signatures during tool replay, a fix for developers using that model workflow.
Why it mattersOther changes include approval prompts for project hooks and GitHub pull-request and issue links in channel events. It is a nightly build, so these are preview changes rather than a stable-release promise. A useful test for anyone already working in Qwen Code, but perhaps not the build to install five minutes before a demo.
Discuss: Would you try a nightly build for practical model and workflow fixes, or wait until they reach a stable release?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.