So, you want to summon the power of Stable Diffusion without selling your soul to Colab notebooks or tangling with cursed Python dependencies? Good news, fellow pixel wizard — with Docker, a Windows 11 PC, and a little help from your NVIDIA GPU, you can run Invoke AI locally without setting your computer on fire. In this guide, we’ll walk you through the DIY PLUG & PLAY approach. No beards need stroking, no YAML sacrifices required.
Step 1: Install Docker Desktop with WSL2 enabled on Windows
Download Docker Desktop at https://www.docker.com/products/docker-desktop and run the installer. When the installer asks you to useWSL2 backend, say yes.
Reboot and log back in.
Now lets enable WSL2, open a Windows Power Shell as Administrator and type:
wsl --install
wsl --update
wsl --set-default-version 2
Step 2: Install Nvidia Drivers and Enable GPU for WSL2
Install the latest drivers for your NVIDIA card, https://www.nvidia.com/Download/index.aspx, studio or game ready.
Step 3: Prepare Your Project Folder
Open a Windows PowerShell and run:
mkdir C:\invokeai-docker
cd C:\invokeai-docker
mkdir C:\invokeai-docker\data
mkdirC:\invokeai-docker\outputs
mkdir C:\invokeai-docker\models
Step 4: Create docker-compose.yml with GPU + Persistent Volumes
Download the file named docker-compose.yml into C:\invokeai-docker:
Step 5: Launch Invoke AI with Docker Compose
In Windows PowerShell run:
cd C:\invokeai-docker
docker compose up -d
This will download and run the container, mount folders for persistent data, open up InvokeAI’s web UI port (9090), and use the GPU for processing if available.
Step 6: Opening InvokeAI webUI
In your browser, go to: http://localhost:9090
You should see the Invoke AI interface ready to use!
Other useful commands:
Stop a docker container: docker compose down
Stop a docker container: docker compose up -d