GitHub radar
Qwen-Image-2.1 for ComfyUI: generate and edit images
Comfy-Org packaged Alibaba's Qwen-Image-2.1 for ComfyUI — a single model for both text-to-image generation and photo editing, with 535 thousand downloads in one month.
Comfy-Org packaged Alibaba's Qwen-Image-2.1 for ComfyUI users, with pre-built workflow templates for both generation and editing. This is a single 7B-parameter model that handles text-to-image generation and photo editing in one workflow. Give it a text description and it produces an image. Load an existing photo, describe what you want changed, and it edits it. The model accepts up to ten reference photos to guide style or subject identity. It also supports transparent (RGBA) backgrounds, which is useful for product and catalog photography. The license is research-only, so commercial use requires checking the terms.
Why a vibe-coder should care
If you already use ComfyUI, this package saves you the setup time. Before this release, generation and editing required different models in separate workflows. Here, one model handles both tasks. 535 thousand downloads in one month show that the ComfyUI community has already adopted it.
How to install
Copy this and send it to your agent — Claude Code, Codex, any of them:
Set up Qwen-Image-2.1 for ComfyUI from this package: https://huggingface.co/Comfy-Org/Qwen-Image-2.1 — download the model files, load a pre-built workflow and show me how to generate the first image from a text description.
Estimated based on parameter count: a GPU with 12 GB of VRAM or a Mac with 24 GB of RAM. Alibaba has not published official hardware requirements.
Open on Hugging Face▌ More finds