ComfyUI logo

ComfyUI

100% Free
#18 in AI Video Tools Last verified: June 12, 2026

ComfyUI is a highly advanced, node-based graphical user interface (GUI) designed to run Stable Diffusion and open video models (like Wan, Hunyuan, LTX, and CogVideo) locally. It offers infinite control over workflows, and has no recurring costs as it runs entirely on your own GPU.

Why this tool isn't working

You may wish to try one of these alternatives.

How to use ComfyUI

1

Download and install

Clone the GitHub repository or download the standalone desktop package for Windows.

2

Download model weights

Download video model files (e.g. Wan 2.2) and place them in the models/checkpoints folder.

3

Load a workflow

Drag and drop an image containing a ComfyUI workflow JSON into the browser workspace.

4

Execute queue

Click "Queue Prompt" to generate the video frames locally using your GPU.

Report Issue

Help us keep our directory accurate.