Stable Diffusion web UI docker build

A web interface for Stable Diffusion. Build based on Ubuntu 22.04, Cuda 12.6.2 & Python 3.10 using latest release from https://github.com/SonycProduction/stable-diffusion-webui.

Features

• Original txt2img and img2img modes
• One click install and run script (but you still must install python and git)
• Outpainting
• Inpainting
• Color Sketch
• Prompt Matrix
• Stable Diffusion Upscale
• Loopback, run img2img processing multiple times
• X/Y/Z plot, a way to draw a 3 dimensional plot of images with different parameters
• Textual Inversion
• and many more...

Installation and Running

Quick start (assuming docker basic setup exists, refer to prerequisites for any issues)

curl -sSL https://raw.githubusercontent.com/SonycProduction/stable-diffusion-webui/refs/heads/master/docker-stable-diffusion-webui.sh | bash

Install stable-diffusion-webui in docker. Files are created in current location in folder stable-diffusion-webui

refer to prerequisite if any issues are encountered

Prerequisite:

1. Docker

Add Docker's apt repository.

# Add Docker's official GPG key:
sudo apt-get update
sudo apt-get install ca-certificates curl
sudo install -m 0755 -d /etc/apt/keyrings
sudo curl -fsSL https://download.docker.com/linux/debian/gpg -o /etc/apt/keyrings/docker.asc
sudo chmod a+r /etc/apt/keyrings/docker.asc

# Add the repository to Apt sources:
echo \
  "deb [arch=$(dpkg --print-architecture) signed-by=/etc/apt/keyrings/docker.asc] https://download.docker.com/linux/debian \
  $(. /etc/os-release && echo "$VERSION_CODENAME") stable" | \
  sudo tee /etc/apt/sources.list.d/docker.list > /dev/null
sudo apt-get update

If issues occur, you may need to substitute the part $VERSION_CODENAME of this command with the codename of the corresponding Debian release, such as jammy or bookworm.

Install Docker (and related packages)

sudo apt install docker-ce docker-ce-cli containerd.io docker-buildx-plugin docker-compose-plugin

Test Docker Installation

sudo docker run hello-world

2. Docker Compose (Optional - Recommended)

sudo apt install docker-compose

docker compose command should already be bundled with Docker, this installs commonly used docker-compose

3. Nvidia Container Toolkit

Add Nvidia Container toolkit repository

curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \
  && curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \
    sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \
    sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list

Update packages from Nvidia container Toolkit repository

sudo apt update

Install Nvidia Container Toolkit

sudo apt install nvidia-container-toolkit

Configure for use with Docker

sudo nvidia-ctk runtime configure --runtime=docker

sudo systemctl restart docker

Test with Docker

sudo docker run --rm --runtime=nvidia --gpus all nvidia/cuda:12.6.0-base-ubuntu22.04 nvidia-smi

Automatic1111 (Stable Diffusion UI) Installation on Docker:

Clone this repo

git clone https://github.com/SonycProduction/stable-diffusion-webui.git

Navigate into the cloned directory

cd stable-diffusion-webui

Option 1 (Build and deploy in a single command - Recommended)

Build and deploy the container in a single command. WebUI is available on port 7860. Required folders are mapped in current location.

sudo docker compose up --build

docker-compose.yml file can be modified to change folder locations and access port for webui. Stable Diffusion WebUI starts with following commandline arguments enabled: --listen --api --medvram --xformers --enable-insecure-extension-access --allow-code --administrator You can change the commandline arguments (e.g. --lowvram --no-half --precision full) depending on your needs.

Option 2 (Manually building image):

1. Build image using Dockerfile

sudo docker build -t <image_name:tag> .

<image_name:tag> change it to user's image name.

2. Run the image built in previous step

sudo docker run -d \
  --name stable-diffusion-webui \
  --restart unless-stopped \
  -p 7860:7860 \
  -v ./models:/app/models \
  -v ./extensions:/app/extensions \
  -v ./embeddings:/app/embeddings \
  -v ./outputs:/app/outputs \
  --runtime nvidia \
  --gpus all \
  --deploy-resources-reservations-devices 'driver=nvidia, count=all, capabilities=[gpu]' \
  <image_name:tag>

<image_name:tag> needs to be replaced the one defined in previous step. This wil start the container and map the required volumes as folders in current location.

It’s done 🎊🎉👏 Go to browser, and type 127.0.0.1:7860 to access WebUI.