Abstract

Recent visual generative models often struggle with consistency during image editing due to the entangled nature of raster images, where all visual content is fused into a single canvas. In contrast, professional design tools employ layered representations, allowing isolated edits while preserving consistency. Motivated by this, we propose \textbf{Qwen-Image-Layered}, an end-to-end diffusion model that decomposes a single RGB image into multiple semantically disentangled RGBA layers, enabling \textbf{inherent editability}, where each RGBA layer can be independently manipulated without affecting other content. To support variable-length decomposition, we introduce three key components: (1) an RGBA-VAE to unify the latent representations of RGB and RGBA images; (2) a VLD-MMDiT (Variable Layers Decomposition MMDiT) architecture capable of decomposing a variable number of image layers; and (3) a Multi-stage Training strategy to adapt a pretrained image generation model into a multilayer image decomposer. Furthermore, to address the scarcity of high-quality multilayer training images, we build a pipeline to extract and annotate multilayer images from Photoshop documents (PSD). Experiments demonstrate that our method significantly surpasses existing approaches in decomposition quality and establishes a new paradigm for consistent image editing. Our code and models are released on this https URL

Paper: https://arxiv.org/abs/2512.15603

Code: https://github.com/QwenLM/Qwen-Image-Layered

Blog: https://qwenlm.github.io/blog/qwen-image-layered/

Hugging Face: https://huggingface.co/Qwen/Qwen-Image-Layered

Demo: https://huggingface.co/spaces/Qwen/Qwen-Image-Layered

Modelscope: https://modelscope.cn/models/Qwen/Qwen-Image-Layered

Comfy-Org files: https://huggingface.co/Comfy-Org/Qwen-Image-Layered_ComfyUI/tree/main

GGUFs: https://huggingface.co/QuantStack/Qwen-Image-Layered-GGUF/tree/main