Skip to Content
Subscribe to my YouTube Channel: @hakehardware
Artificial IntelligenceComfyUI + Ubuntu Server + CUDA 13 + PyTorch 2.9 (NVIDIA)

Install ComfyUI on Ubuntu Server 24.04

comfyui

Install the NVIDIA drivers

Secure Boot MUST be turned off in BIOS

Update your system:

sudo apt update && sudo apt upgrade -y

Add NVIDIA PPA

sudo add-apt-repository ppa:graphics-drivers/ppa && sudo apt update

Check for available drivers

ubuntu-drivers devices

Install drivers (I usually choose the “recommended” option)

sudo apt install nvidia-driver-<DRIVER VERSION>

Reboot

sudo reboot now

Install Dependencies

The following should be installed:

sudo apt install python3.12 python3.12-venv python3-pip git build-essential libgl1 libglib2.0-0 -y

Install ComfyUI

Clone the repo (I usually just clone it to my home folder)

git clone https://github.com/comfyanonymous/ComfyUI.git

Enter the folder

cd ComfyUI

Create a virtual environment

python3.12 -m venv venv

Enter the virtual environment

source venv/bin/activate

Now install PyTorch

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

Now install the requirements

pip install -r requirements.txt

This will install ComfyUI.

Run ComfyUI

To start ComfyUI run:

python main.py --listen 0.0.0.0

Now on a separate computer, navigate to the IP of the PC running ComfyUI with port 8188

http://192.168.10.4:8188/

And that’s it!

Last updated on