8000 GitHub - alyssonw2/ComfyUI-Installer: Easily install ComfyUI on a Linux system. Installs completely inside a Python venv.
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

alyssonw2/ComfyUI-Installer

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ComfyUI Installer

Origional Author: https://github.com/HAMM3REXTREME/ComfyUI-Installer Modified Guided installer.sh by: https://github.com/itsdarklikehell

Easily install ComfyUI + ComfyUI-Manager + ComfyUIMini (in a python venv) on Linux.

Tested on Arch + AMD GPU (by https://github.com/HAMM3REXTREME).

Tested on Ubuntu + Nvidia GPU (https://github.com/itsdarklikehell).

ComfyUI Screenshot Note: This is not the official ComfyUI icon.

Installs ComfyUI, ComfyUI Manager and ComfyUIMini in a python venv

A Quick, Guided install (no .settings file present, it will create one) (SHOULD ALSO WORK ON ALL DISTROS)

You will need to have python, python-venv and pip on your system. Make sure to install for your GPU Vendor (AMD/NVIDIA):

git clone https://github.com/itsdarklikehell/ComfyUI-Installer && cd ComfyUI-Installer && ./install.sh

A Quick, Unattended/headless install (i.e one where there is a .settings file present) (SHOULD ALSO WORK ON ALL DISTROS)

You will need to have python, python-venv and pip on your system. Make sure to install for your GPU Vendor (AMD/NVIDIA):

git clone https://github.com/itsdarklikehell/ComfyUI-Installer && cd ComfyUI-Installer

Create a .settings file in the ComfyUI-Installer folder containing the following:

# The directory where the installer is located:
export COMFYUI_INSTALLER_DIR=~/ComfyUI-Installer
# The directory where the ComfyUI is located:
export COMFYUI_DIR=~/ComfyUI-Installer/ComfyUI
# The type of GPU to use:
export GPU=NVIDIA
# The directory where the backups are located:
export BACKUP_DIR=~/ComfyUI-Installer/backup
# The virtual environment directory:
export VIRTUAL_ENV=~/ComfyUI-Installer/ComfyUI/venv
# Use systemd:
export USE_SYSTEMD=true

Then run the install script:

./install.sh

Detailed explation of the installation process:

  1. To install ComfyUI using this script, clone this repo and cd into it: git clone https://github.com/itsdarklikehell/ComfyUI-Installer && cd ComfyUI-Installer

  2. Then run the install script: ./install.sh

    Upon first run, if there is no .settings file present it will ask you for some information and it stores that in a .settings file like this:

    # The directory where the installer is located:
    export COMFYUI_INSTALLER_DIR=~/ComfyUI-Installer
    # The directory where the ComfyUI is located:
    export COMFYUI_DIR=~/ComfyUI-Installer/ComfyUI
    # The type of GPU to use:
    export GPU=NVIDIA
    # The directory where the backups are located:
    export BACKUP_DIR=~/ComfyUI-Installer/backup
    # The virtual environment directory:
    export VIRTUAL_ENV=~/ComfyUI-Installer/ComfyUI/venv
    # Use systemd:
    export USE_SYSTEMD=true

    If do you have a .settings file already present, it will just use that and then proceeds with the installation.

  3. When the install script has finished, if you have chosen to use a systemd service, ComfyUI should now be starting/running, you can check the status of the service with: sudo systemctl status ComfyUI.service If it's not running, you can start it with: sudo systemctl start ComfyUI.service If you want to enable it to start on boot, you can do so with: sudo systemctl enable ComfyUI.service

    If you have chosen not to use systemd, you can start ComfyUI with one of the during installation created run scrips: ./scripts/run_gpu.sh or ./scripts/run_cpu.sh.

If everyting has worked, ComfyUI, ComfyUI-Manager and ComfyUIMini should be installed. Open your browser and go to: http://0.0.0.0:8188/ to view ComfyUI's interface (Works better on desktop devices/browsers). http://0.0.0.0:3000/ to view ComfyUIMini's interface (Should work better with mobile devices/browsers). To check ComfyUI's status, run: tail -f ComfyUI/logs/comfyui.log or journalctl -f -u ComfyUI.service To check ComfyUIMini's status, run: tail -f ComfyUI/custom_nodes/ComfyUIMini/logs/comfyuimini.log or journalctl -f -u ComfyUIMini.service

Usage

To Manually launch ComfyUI use:

./scripts/run_gpu.sh or ./scripts/run_cpu.sh.

To Manually launch ComfyUIMini use:

./ComfyUI/custom_nodes/ComfyUIMini/scripts/start.sh.

To start ComfyUI systemd service use:

sudo systemctl start ComfyUI.service

To start ComfyUIMini systemd service use:

sudo systemctl start ComfyUIMini.service

To enable ComfyUI systemd service use:

sudo systemctl enable ComfyUI.service

To enable ComfyUIMini systemd service use:

sudo systemctl enable ComfyUIMini.service

To stop ComfyUI systemd service use:

sudo systemctl stop ComfyUI.service

To stop ComfyUIMini systemd service use:

sudo ystemctl stop ComfyUIMini.service

To restart ComfyUI systemd service use:

sudo systemctl restart ComfyUI.service

To restart ComfyUIMini systemd service use:

sudo systemctl restart ComfyUIMini.service

To check ComfyUI systemd service status use:

sudo systemctl status ComfyUI.service

To check ComfyUIMini systemd service status use:

sudo systemctl status ComfyUIMini.service

Updating

Updating ComfyUI

Simply re-run the install.sh script, it should detect that ComfyUI is already installed and then proceed to try and update it. Or if it is already running you could use the ComfyUI-Manager to update everything.

Troubleshooting

If you get the "Torch not compiled with CUDA enabled" error

Uninstall torch with:

pip uninstall torch

And install it again with the command (for Nvidia) above.

For AMD cards not officially supported by ROCm

Try running it with this command if you have issues:

For 6700, 6600 and maybe other RDNA2 or older: HSA_OVERRIDE_GFX_VERSION=10.3.0 python main.py

For AMD 7600 and maybe other RDNA3 cards: HSA_OVERRIDE_GFX_VERSION=11.0.0 python main.py

You can add these changes (and other args you want) to scripts/run_gpu.sh for convenience.

About

Easily install ComfyUI on a Linux system. Installs completely inside a Python venv.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Shell 100.0%
0