Picture published by the creator on GitHub.
ComfyUI: Build image, video and audio AI workflows by linking nodes
Added · Updated
Made by Comfy Org
View profile on GitHub (opens in a new tab)Build image, video, audio and 3D generation workflows by connecting blocks on a graph; for creators who want control, on Windows, Linux or macOS.
What it does
ComfyUI is a node graph program: each step of an AI creation process is a block, and you connect blocks to make a workflow for images, video, audio, 3D and text without writing code. It focuses on local consumer hardware, with queueing, partial re-running of a graph and support for quantized (compressed) models.
How to use it
Input: a workflow, either one you connect yourself or a ready-made template from the project's workflow gallery. Action: run it in the desktop app, a manual install, or the cloud version. Output: the generated image, video, audio or 3D result. App Mode can turn a complex workflow into a simple screen.
Common use cases
- Pick a text to image template, change the prompt and settings, and get a generated picture.
- Share a workflow you built as a simple App Mode screen for someone who does not want the graph.
- Connect a workflow to your own software through the API so it produces results in a production pipeline.
Getting started
Reviewed setup route · source checked 2026-10-21 · Source page (opens in a new tab)
- Download the desktop application for Windows or macOS, which the source calls the easiest way to start.
- Open it and load a template workflow from the source’s workflow library.
- Press Ctrl+Enter to queue the workflow and generate.
- Try with
- Start from a ready-made template workflow rather than building a node graph from scratch.
- What to expect
- A generated image, video or audio result produced from the node graph you ran; workflows can be saved and reloaded.
- Access note
- The desktop app covers Windows and macOS; the manual install is the route for other systems and GPU types (NVIDIA, AMD, Intel, Apple Silicon, Ascend). The source also lists a separate hosted Comfy Cloud version for people without local hardware; check its page for terms.
Can I use this?
Requirements Most people start with the desktop app on Windows or macOS. Manual installation is the terminal route and supports NVIDIA, AMD, Intel, Apple Silicon and Ascend GPUs. The source says large models can run on as low as 4GB VRAM + 8GB RAM; that is the project's claim.
Caveat The source's statements about speed and model support are the project's own claims. Which models work well on your hardware is not documented beyond that memory figure.
Worth knowing The source says its official cloud version is paid and is separate from running it locally. Partner nodes give access to closed source models; their pricing and account needs are not documented. Licensed under GNU General Public License v3.0.
Why it's here
Not trending this week
Last seen on the GitHub trending list on 26 October 2026. It stays on BuildTube so you can still find it.
- 145,710 people have starred it on GitHub.
Numbers checked on 8 November 2026; not refreshed since.
Behind it
View setup guideBuildTube has not run or verified this project. Everything above is written from what the creator published.
Made this?
Quick questions
- What is ComfyUI?
- ComfyUI is a node graph program: each step of an AI creation process is a block, and you connect blocks to make a workflow for images, video, audio, 3D and text without writing code.
- What can I use ComfyUI for?
- Pick a text to image template, change the prompt and settings, and get a generated picture; Share a workflow you built as a simple App Mode screen for someone who does not want the graph; Connect a workflow to your own software through the API so it produces results in a production pipeline
- Can I run ComfyUI on my computer?
- It installs as a desktop app on your own computer. Most people start with the desktop app on Windows or macOS. Manual installation is the terminal route and supports NVIDIA, AMD, Intel, Apple Silicon and Ascend GPUs. Which models work well on your hardware is not documented beyond that memory figure.
- What do I need to use ComfyUI?
- A computer running Windows, Linux or macOS, or the project's cloud version. The source says it supports NVIDIA, AMD, Intel, Apple Silicon and Ascend GPUs; storage and model download needs are not documented here.
- What license does ComfyUI use?
- Licensed under GNU General Public License v3.0.
More from this week
- Spec Kit Turn a written spec into plans, tasks and code
Spec Kit gives your AI coding agent a step-by-step process to build a feature, fix a bug or assess an idea.
- RAGFlow Ask questions about your documents and get cited answers
Upload documents and ask questions with traceable citations.
- P-H3-Image-Comfy Make and edit 8-megapixel images in ComfyUI
Turn short text requests into 8-megapixel images and edit pictures on your own graphics card.
Explore by topic
Post this
Download share imageFor makers
Made this? Show it’s on BuildTube — add the badge to your README or embed the card on your site.
README badge (Markdown)
[](https://buildtube.app/p/gh-comfy-org-comfyui/)Embed card (HTML)
<iframe src="https://buildtube.app/embed/gh-comfy-org-comfyui/" width="560" height="240" style="border:0;max-width:100%" loading="lazy" title="ComfyUI on BuildTube"></iframe>