selfhost.directory

The index of self-hostable projects — ranked, categorized and served with live updates.
Own your data and stop giving it to big tech.

ComfyUI logo

Self-host ComfyUI

v0.45.0

Modular AI engine for visual content with node graph

ReleaseOpen SourceGPL-3.0
Updated5 days ago
LanguagePython
Setup difficultyModerate
Stars
147 014
RAM
8GB
Port
8188 (Web UI and API server)
Install with Linux

What is ComfyUI?

ComfyUI is a self-hosted AI engine for visual content creation built around a visual node graph interface. It generates various media types using modern diffusion models. The software runs locally on Windows, Linux, and macOS with support for NVIDIA, AMD, Intel, Apple Silicon, and Ascend hardware. Key capabilities include asynchronous queueing, intelligent VRAM and RAM handling, model offloading, quantized model support, custom nodes, workflow templates, App Mode, and a local API for integration. Users can install it via a desktop app, portable builds, or manual Python setup.

Should you self-host ComfyUI?

ComfyUI scores 92/100 on selfhost.directory (how scoring works) and ranks #2 of 123 in AI. It has 147 014 GitHub stars, +10 054 in the last 30 days.

It shipped 29 stable releases in the past year. The latest, v0.45.0, came out on 10 November 2026. New versions land roughly every 6 days.

Setup is moderate: expect some configuration beyond a single command. It is GPL-3.0-licensed and written mostly in Python.

Its closest alternative on selfhost.directory is openclaw (93/100, 394 525 stars). It scores higher overall, so compare the two before you pick. See ComfyUI vs openclaw side by side.

Features

  • App Mode
  • Broad native model support
  • Built-in media processing tools
  • Custom nodes extension system
  • Dynamic prompts
  • Efficient local execution
  • High bit depth image/video support
  • Local API for integration
  • Model format flexibility
  • Multi-modal generation
  • Node-based workflow editor
  • Offline Mode
  • Reusable subgraphs and workflow templates
  • Textual inversion/embeddings support
  • Workflow save/load as JSON

ComfyUI README

ComfyUI

The most powerful and modular AI engine for content creation.

Website Dynamic JSON Badge Twitter Matrix

ComfyUI Screenshot

How to install ComfyUI

Install ComfyUI with Linux

Manual Install (Windows, Linux)

Python 3.14 works but some custom nodes may have issues. The free threaded variant works but some dependencies will enable the GIL so it's not fully supported.

Python 3.13 is very well supported. If you have trouble with some custom node dependencies on 3.13 you can try 3.12

torch 2.7 is minimally supported but using a newer version is extremely recommended. Using a cu130 or above version of pytorch is required on Nvidia 20 series and above. Some features and optimizations might only work on newer versions. We generally recommend using the latest major version of pytorch with the latest cuda version unless it is less than 2 weeks old. If your pytorch is more than 6 months old, please update it.

Instructions:

Git clone this repo.

Put your SD checkpoints (the huge ckpt/safetensors files) in: models/checkpoints

Put your VAE in: models/vae

AMD GPUs (Linux)

AMD users can install rocm and pytorch with pip if you don't have it already installed, this is the command to install the stable version:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/rocm7.2

This is the command to install the nightly with ROCm 7.2 which might have some performance improvements:

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/rocm7.2

Intel GPUs (Windows and Linux)

Intel Arc GPU users can install native PyTorch with torch.xpu support using pip. More information can be found here

  1. To install PyTorch xpu, use the following command:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/xpu

This is the command to install the Pytorch xpu nightly which might have some performance improvements:

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/xpu

NVIDIA

Nvidia users should install stable pytorch using this command:

pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu130

This is the command to install pytorch nightly instead which might have performance improvements.

pip install --pre torch torchvision torchaudio --index-url https://download.pytorch.org/whl/nightly/cu132

Dependencies

Install the dependencies by opening your terminal inside the ComfyUI folder and:

pip install -r requirements.txt

After this you should have everything installed and can proceed to running ComfyUI.

Manual install with PyTorch for NVIDIA (CUDA), AMD (ROCm), Intel (XPU); requires Python 3.10+, PyTorch 2.7+

How to configure ComfyUI

  • --listen127.0.0.1IP address to listen on (default: 127.0.0.1)

    values: IP address or comma-separated list, or empty for 0.0.0.0,::

  • --port8188Listen port

    values: Port number

  • --tls-keyfilePath to TLS (SSL) key file

    values: File path

  • --tls-certfilePath to TLS (SSL) certificate file

    values: File path

  • --enable-cors-headerEnable CORS with optional origin

    values: Origin or * for all

ComfyUI releases & version history

@ Every self-hosted release, weekly

One Monday email with the week's stable releases across the directory.

Models ComfyUI runs

Popular open-weight models with ComfyUI support, with VRAM needs and licence at a glance.

Qwen logo7.1BNon-commercial≈8.9 GB↓ 274K
Tongyi-MAI logo
Z-Image-TurboTongyi-MAI
6.2BCommercial OK≈7.8 GB↓ 775.7K
black-forest-labs logo
FLUX.1-devblack-forest-labs
11.9BNon-commercial≈15 GB↓ 476K
Qwen logo7.1BNon-commercial≈8.9 GB↓ 37.4K
black-forest-labs logo
FLUX.1-schnellblack-forest-labs
11.9BCommercial OK≈15 GB↓ 276.8K
Qwen logo20.4BCommercial OK≈25 GB↓ 353.3K
krea logo12.8BConditions≈16 GB↓ 85.6K
OpenMOSS-Team logo
MOSS-Image-SFTOpenMOSS-Team
5.1BCommercial OK≈6.5 GB↓ 23.8K

Browse all 206 models →

More in AI

Rate this project

Click to rate — no account needed.

0 comments

Posting as guest
Loading comments…