Designers and developers who want to edit one element of a picture without redrawing the rest get free Apache 2.0 weights.
Kenning's take
What it does
Qwen-Image-Layered breaks a picture into several RGBA layers, so one element can be moved, resized or recoloured without touching the rest. It builds on Qwen-Image and was released with open weights under Apache 2.0 in December 2025. It runs through the diffusers library, with a Hugging Face demo.
- Decomposes images into RGBA layers
- Edit, resize, move or recolour one layer
- Built on Qwen-Image
- Apache 2.0 open weights
- Runs with Hugging Face diffusers
Who it suits. Designers and developers who want layer-based editing of generated or existing images.
Key facts
- Category
- AI models
- Part of
- Qwen
- Made by
- Alibaba (Qwen team)
- Based in
- China
- Released
- 17 December 2025
- How to get it
- Free download from Hugging Face or ModelScope; online demo on Hugging Face Spaces
- Works on
- self-hosted, Hugging Face
- In the UK
- Yes
- Free plan
- Yes, open weights
FAQ
Is Qwen-Image-Layered free?
Yes. The weights are open under Apache 2.0.
What does Qwen-Image-Layered do?
It splits an image into separate transparent layers that can be edited one at a time.
Facts from the maker's own pages and the sources listed. Checked 2026-10-11.
Similar tools
MAI-UI
Alibaba Tongyi research models that operate smartphone apps by looking at the screen and tapping
Qwen-Image-2.0
Alibaba's 7-billion-parameter model that both creates and edits images, with strong text rendering at 2K
Qwen-Image-3.0
Alibaba's hosted image model built for text-heavy layouts, taking prompts up to 4,500 tokens
Qwen VLo
Preview multimodal model from Alibaba that understands images and generates or edits them from chat prompts