Fork of the amazing work of https://github.com/ashleykleynhans/stable-diffusion-docker but unfortunatley no longer available.
Welcome! This container will (hopefully) get you started on custom image generation tools!
Contains everything to start your image generation:
- A1111 Stable Diffusion: Simplified image generation. Perfect for beginners.
- ComfyUI: Workflow based /config based and usually uses latest technology. More advanced.
- InvokeAI: ComfyUI alternative.
Additional software:
- JupyterLab: Notebook interface to upload / download files / modify.
- Models / LORAs / etc are synced between all AI generational tools, download on Stable Diffusion, and it will appear under ComfyAI.
- A1111 Stable Diffusion contains a set of plugins to help you get started.
- LINUX ONLY. Does not run on macOS / Windows through Podman Machine.
- NVIDIA GPUs ONLY.
- AMD64 ARCHITECTURE ONLY.
- Does not include ANY SAFETENSORS. Add your models through Stable Diffusion Web UI, or by manually uploading them via Jupyter Lab.
- Made slim as possible, some "default" models such as SDXL are removed. Download them manually.
Recommended to use Linux as there are issues building this container with Podman Machine on macOS / Windows. The reasoning is that pip and other tools used to initially build the image rely being on AMD64 / Linux architecture.
# Clone
git clone git clone https://github.com/cdrage/stable-diffusion-podman.git
cd stable-diffusion-podman
# See build-args.env for all environment variables regarding versions / you can customise your own version
podman build -t stable-diffusion-podman --build-arg-file build-args.env .podman run -d \
--gpus all \
-v /workspace \
-p 3000:3001 \
-p 3020:3021 \
-p 8000:8000 \
-p 8888:8888 \
-p 9090:9090 \
-e JUPYTER_PASSWORD=Jup1t3R! \
stable-diffusion-podman| Connect Port | Internal Port | Description |
|---|---|---|
| 3000 | 3001 | A1111 Stable Diffusion Web UI |
| 3020 | 3021 | ComfyUI |
| 8000 | 8000 | Application Manager |
| 8888 | 8888 | Jupyter Lab |
| 9090 | 9090 | InvokeAI |