TREE NEWS update: Alibaba’s Qwen team open-sourced the image model Qwen-Image-2.1, which unifies text-to-image generation and image editing in a single model with a 7B-parameter visual generation component. The model natively supports transparent image generation and editing, accepts up to 10 reference images for multi-image composition, and adds local editing tools such as lasso selection and brushing. It also improves fidelity in portrait and product editing and text layout.
Alibaba’s Qwen Team Open-Sources Image Model Qwen-Image-2.1
The notable shift here is architectural: folding generation and editing into one 7B model, rather than shipping separate tools, lowers the integration burden for developers building image pipelines. Transparent output and multi-reference composition matter most for product and design workflows, where compositing and cutouts are routine. The open-source release also keeps Qwen competitive with other openly licensed image stacks, which matters for teams wary of closed APIs. Whether native transparency and lasso-style local edits hold up under real production use, rather than demos, is the open question.
Generated by AI for reference only.
Share on WeChat
Open WeChat → Scan → then tap "…" to send to a chat or Moments.
Tap "…" in the top-right corner to send to a chat or share to Moments.