# QwenLM/Qwen-Image-Layered

Qwen-Image-Layered: Layered Decomposition for Inherent Editablity

Repository: https://github.com/QwenLM/Qwen-Image-Layered
Canonical: https://ross.abutalabs.com/products/qwen-image-layered
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2025-12-31T11:40:35+00:00

## Health v2 (maintenance only)
Score: 43/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 60, release rhythm 35, longevity 18
- inputs: {"age_days": 258, "days_push": 245, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2079, forks 170 (observed 2026-08-28T04:06:11.509880+00:00)

## What it is
Qwen-Image-Layered is a diffusion-based model and pipeline that decomposes an input image into multiple independently editable RGBA layers. Each layer can be resized, repositioned, or recolored without affecting other content, enabling high-fidelity image editing.

## Use cases
- decompose an image into editable RGBA layers
- reposition or resize objects in a photo without affecting the background
- recolor isolated image elements independently
- edit occluded or layered image components
- run image-to-layered-decomposition inference with diffusers

## When to choose
- you need layer-based, non-destructive image editing driven by a generative model
- you want to isolate semantic components of an image into separate RGBA layers
- you already use Hugging Face diffusers and want a drop-in pipeline

## When to avoid
- you need text-to-image generation of layered output - the released weights are tuned only for image-to-layer decomposition
- you need fine-grained control over the semantic content of individual layers via prompts
- you lack a GPU or cannot run large diffusion models

## Facets
- artifact type: library
- maturity: experimental
- function: image-processing, machine-learning, deep-learning
- domain: image-processing, artificial-intelligence, graphics
- platform: python, cross-platform
- tags: image-decomposition, rgba-layers, diffusion-model, image-editing, diffusers, qwen, gpu

## Member repositories
- QwenLM/Qwen-Image-Layered (main) score 43

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:11.509880+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T02:56:03.628570+00:00, confidence not recorded.
  - readme: https://github.com/QwenLM/Qwen-Image-Layered (fetched 2026-08-28T04:06:11.509880+00:00, sha de9ecf585965)
- Data as of 2026-08-30T08:39:29.467469+00:00.