Stable Diffusion
Getting started
Stable Diffusion for Beginners
- Part 1: Getting Started: Overview and Installation
- Part 2: Stable Diffusion Prompts Guide
- Part 3: Stable Diffusion Settings Guide
- Part 3: Models
- Part 4: LoRAs
- Part 5: Embeddings/Textual Inversions
- Part 6: Inpainting
- Part 7: Animation
More links
- Stable Diffusion prompt: a definitive guide
- Beginner’s guide to inpainting
- How to write an awesome stable diffusiion prompt
- How to remove extra limbs with Stable Diffusion inpainting
- Overview of popular models
- Stable Diffusion WebUI Settings Explained – Beginners Guide
- NextDiffusion.ai has detailed tutorials and descriptions of features
Installing Easy Diffusion
This is probably the most common
- Easy Diffusion by Cmdr2: https://github.com/cmdr2/stable-diffusion-ui
- other(?) Easy Diffusion repository
- a.k.a.: "Automatic1111"
Starting Stable Diffusion and first run
- a
run.batfile should be in the install directory. Execute it and after it has finished loading, a web interface should come up - type in some text in the main prompt box and click the "generate" button
- e.g. "mountains at sunset"
Configuration and advanced usage
Stable Diffusion has a LOT of features by default and even more as you download models, model modifiers, and feature plugins
Adding models and other enhancements
where various model files go:
- Models/checkpoints go in the stable-diffusion-webui/models/Stable-diffusion folder (.ckpt or .safetensors).
- LoRa/hypernetworks/VAE go in their corresponding /models/folder (folders will be created after you run the install for the first time).
- Upscaling models go in /models/ESRGAN (.pth file).
- Textual embeddings/negative embeddings go in the stable-diffusion-webui/Embeddings folder (.pt file).
- VAEs go in the
models/VAEfolder
Fine-tuning the output
Using VAEs
See this tutorial by NextDiffusion.ai