Skip to content

GETTING_STARTED.md

Latest commit

 

History

History
85 lines (54 loc) · 2.89 KB

GETTING_STARTED.md

File metadata and controls

85 lines (54 loc) · 2.89 KB

Getting Started with a1111-docker

This guide is for anyone new to Docker or AUTOMATIC1111 (A1111). It explains the basics in plain language and points you back to the main README.md for deeper technical details.

🐳 What is Docker?

Docker is a way to package software into containers.
Think of a container as a lightweight, portable box that includes everything the program needs (libraries, dependencies, configs).
The benefit: the software runs the same way on any machine, without you having to manually install all the pieces.

🎨 What is AUTOMATIC1111 (A1111)?

AUTOMATIC1111 is a popular web interface for Stable Diffusion.
It lets you generate, edit, and experiment with AI-generated images through a browser.
This build runs A1111 inside a Docker container, so you don't have to worry about CUDA, Python, or dependency setup.

📦 Why this build?

This Docker setup separates what's ephemeral (rebuilt each time) from what's persistent (your data and settings):

  • Ephemeral (inside the container):

    • Python environment
    • Installed packages
    • The container filesystem itself

  • Persistent (mounted from your host):

    • models/ → your Stable Diffusion checkpoints
    • outputs/ → generated images
    • configs/ → YAML configs
    • extensions/ → installed extensions
    • embeddings/, logs/, cache/, repositories/
    • UI state files: config.json, ui-config.json, styles.csv

This means you can rebuild or update the container at any time without losing your models, outputs, or settings.

🚀 How to run

See the README.md for full setup instructions.
In short: run the provided setup.sh script, then open your browser to:

http://localhost:7860

That's the A1111 WebUI.

🧩 Adding models

This build expects either Stable Diffusion 1.5 or Stable Diffusion XL (SDXL) models.

  1. Download a model checkpoint (.safetensors or .ckpt) from a trusted source such as Civitai.

  2. Place it in one of these folders (on your host machine):

    models/Stable-diffusion/sd15/ models/Stable-diffusion/sdxl/

  3. Restart the container (docker compose up) and the model will appear in the WebUI dropdown.

🪟 Accessing folders from Windows

If you're running Docker Desktop on Windows:

  • The a1111-docker folder on your host is mounted into the container.
  • That means you can drop models into a1111-docker/models/Stable-diffusion/... directly from Windows Explorer.
  • Outputs will appear in a1111-docker/outputs/ on your host.

✅ Summary

  • Docker runs A1111 in a clean, reproducible environment.
  • Your models, outputs, configs, and extensions are persistent on your host.
  • Add SD1.5 or SDXL models under models/Stable-diffusion/ to get started.
  • Access the UI at http://localhost:7860.
  • For full details, see the main README.md.