Tools for
Conviviality

Home / AI image tools / ComfyUI

ComfyUI

A node-based tool for building image, video and 3D generation workflows that run on your own machine.

Instead of Hosted AI image tools and Adobe Firefly

M · SystemOpen sourceFreeWorks offlineSelf-hosted
www.comfy.org
Screenshot of the ComfyUI website

Screenshot from www.comfy.org, taken 2026-10-23, shown to help you recognise the tool. It belongs to its maker. Ask us to remove it.

A guitarist's pedal board is a chain of small boxes, each doing one thing to the sound, and ComfyUI is a pedal board for making images. It lets you build image, video, audio and 3D generation workflows as a visual graph of nodes. It is a GPL-3.0 project from Comfy Org that runs on Windows, Mac and Linux, and its core runs fully offline and downloads nothing unless you ask.

You choose the models, see every step and keep the workflow as a file. Optional paid partner API nodes can be switched off, and Comfy also sells a paid cloud version for people without suitable hardware. ComfyUI scores a 5 on lock-in and longevity, a 4 on self-direction and power and a 3 on footprint and delivery.

It needs a capable computer, setup takes effort, and each model you load comes with its own licence and its own questions about training data that the score does not cover.

    What stands out

  • Builds image, video and 3D workflows as a visual graph of nodes
  • The core runs fully offline and you choose the models
  • Needs a capable computer and some setup effort
Visit ComfyUI
Do you make ComfyUI? Get the badge for your site

This badge reports a dated review. It is not a certification and it costs nothing. Link it to this page and leave the numbers alone: the image always shows the current score.

For dark backgrounds, use comfyui-dark.svg in place of comfyui-light.svg.

For comparison, we also reviewed Adobe Firefly (15/30) and Midjourney (14/30) as a benchmark.

Other tools for “make images with ai on my terms”