The following steps are required to use Qwen-Image in ComfyUI:
- Download or update ComfyUI: Download the latest version from the ComfyUI website or update an existing installation, making sure that Python 3.8 or later is installed on your system.
- Installation of dependencies: Install the necessary Python libraries via pip, including
diffusersIt is recommended to install PyTorch with CUDA support to improve GPU performance. - Getting model weights: After selecting the Qwen-Image workflow in ComfyUI, you will be prompted to download the model weights automatically, or you can download them manually from Hugging Face or ModelScope.
- Configuration environment: High-performance GPUs (e.g. RTX 4090D, 24GB RAM) are recommended, and the first run takes about 86% of video memory and takes 94 seconds.
- Loading workflows: Select the Qwen-Image workflow in ComfyUI, set the prompt word and parameters (e.g. resolution, inference steps, CFG scale, etc.), and click Run to generate the image.
For details, please refer to ComfyUI official documentation.
This answer comes from the articleQwen-Image: an AI tool for generating high-fidelity images with accurate text renderingThe
May not be reproduced without permission:AI productivity tools " How to install and use Qwen-Image in ComfyUI?
English 
简体中文 
日本語 
Deutsch 
Português do Brasil