comfyanonymous/ComfyUI

ComfyUI

Node-based image and video generation with open models.

Open sourceActiveAn afternoon

“The most powerful and modular diffusion model GUI, api and backend with a graph/nodes interface. The fastest local inference engine in the world.” — from the project

The license, in plain English

License: GPL 3.0 · OSI-approved
Use it at work?: Yes — run it, modify it, sell services around it.
Share your changes?: Only if you distribute the software; then your changes must be GPL too. Running it on your own servers is fine.

Our reading, not legal advice — the repository’s LICENSE file is the source of truth.

Alternatives

AUTOMATIC1111

Stable Diffusion web UI

The classic browser UI for Stable Diffusion models.

Open source★ 165,171 ollama

Ollama

Run open-weight models locally with one command.

Open source★ 185,220 langgenius

Dify

Build LLM apps and agents with a visual workflow.

License unclear★ 159,908 open-webui

Open WebUI

A ChatGPT-style interface for local and remote models.

License unclear★ 157,146 ggml-org

llama.cpp

LLM inference in C/C++ on almost any hardware.

Open source★ 132,752 browser-use

Browser Use

Let AI agents operate websites.

Open source★ 119,469