ComfyUIv0.46.0
A modular node-based GUI for running diffusion models locally to generate images.
148k
Stars
1420.7k
Total downloads
10
Versions
6h ago
Page synced
Download
v0.46.02026-11-18 · 3d agoRecommended for most users
ComfyUI_windows_portable_intel.7z
1.4 GBx643.3k downloadsSHA256 db902d46c7…This is the right pick for most people, including when you are unsure about x64 vs arm64.
Windows — all files
ComfyUI_windows_portable_amd.7z
1.6 GB8.6k downloadsSHA256 00f42ed583…ComfyUI_windows_portable_nvidia_arm64.7z
1.8 GBarm64937 downloadsSHA256 952b8fd26f…ComfyUI_windows_portable_nvidia_cu126.7z
1.8 GB17.2k downloadsSHA256 06ef206970…ComfyUI_windows_portable_nvidia.7z
1.9 GB81k downloadsSHA256 6a5255c7ec…About this project
ComfyUI is a desktop tool for generating images with diffusion models on your own computer. Instead of a single fixed form, it presents the whole generation pipeline as a graph of nodes — loading a model, writing prompts, sampling and decoding become blocks you can freely connect and rearrange. It also offers an API and backend, so workflows can be driven from other tools.
It suits creators and AI art enthusiasts who want fine-grained control over how each image is produced. After installing it, you build a workflow on the node canvas, point it at your model files and start generating. Workflows can be saved, shared and reused, making it easy to reproduce a look or tweak one step at a time.
Key features
- Graph/node interface for building image pipelines
- Runs diffusion models locally
- Modular design that's easy to extend
- API and backend included
- Save and reuse workflows
Use cases & who it's for
- Generating AI images on your own computer without being limited by fixed controls
- Producing images in batches with a workflow you can save and reuse
- Tweaking an AI art pipeline step by step to dial in the result
- Plugging image generation into your own tools or scripts
Maintenance
Actively maintained
Last release 3d ago
License
GPL-3.0
Copyleft license
Allows:Commercial use、Modification、Distribution
Requires:Disclose source、Keep the copyright notice、Same license
Facts
- Language
- Python
- Forks
- 17.6k
- Latest release
- 2026-11-18
- Created
- 2023-01-17
System requirements & installation
- Supported systems
- Windows (based on the files the latest release actually ships)
- How to install
Windows: Download the .exe or .msi installer, run it and follow the setup wizard.
- Which file to pick
Files with "Portable" in the name need no installation — unzip and run.
FAQ
Is ComfyUI free?
Yes. ComfyUI is free and open source under the GPL-3.0 license, and you can download and use it for free.
Which systems does ComfyUI support?
This site provides the Windows build of ComfyUI. Download the matching file from the version list to use it on Windows.
How do I download and install ComfyUI from this site?
Pick the version you want from the version list on this page, download the file for your system, then run the installer or follow the included instructions to start it.
How is ComfyUI different from other AI image tools?
Instead of fixed controls, ComfyUI organizes generation as a graph of nodes, so every step can be connected and adjusted freely — ideal if you want detailed control over the result.
Older versions 9 versions
Related software
An AI assistant that really gets things done for you
The open source coding agent.
Run large language models like DeepSeek and Qwen on your own computer
All-in-one desktop assistant for Claude Code, Codex, OpenCode and more
