Return to course: Stable Diffusion – Level 1 Stable Diffusion Art

Previous Lesson Previous Next Next Section

Stable Diffusion - Level 1

Introduction - Generate your first image
What is Stable Diffusion?
Setting up AUTOMATIC1111
Generate your first image
Using a Checkpoint model to change image style
How to building a good prompt
A list of good keywords
Prompt examples
Recap - Generate your first image

Introduction - text-to-image settings
Prompt
Negative prompt
CFG Scale
Image size
Batch size and batch count
Sampling method
Sampling steps
Seed
Hi Res Fix
Recap - text-to-image settings

Quiz - text-to-image settings

Introduction - Prompt Building
Subject
Medium
Style
Artist
Website
Resolution
Additional details
Color
Lighting
Prompt building process
Exercise - Prompt Building

What are checkpoint models?
How are models made?
Installing a checkpoint model
Popular checkpoint models
Stable Diffusion v2 models

Quiz - checkpoint models

Introduction - Basic inpainting
A simple inpainting workflow
Inpainting settings explained
Inpainting models

Quiz - Basic inpainting

Introduction - SDXL model
What is the SDXL model?
Installing the SDXL model
Using the SDXL models
Improvements in SDXL models
Tips on using the SDXL models

Quiz - SDXL models

Introduction - Fixing common issues
Duplicate images
Not showing full body
Garbled faces
Messed-up fingers

Quiz - Fixing common issues

Introduction - How Stable Diffusion work 1
Diffusion models
How training is done
Reverse diffusion
The Stable Diffusion model
Text-to-image step-by-step

Quiz - How Stable Diffusion work 1

Introduction - End-to-end workflow 1
Step 1: Refine the prompt
Step 2: Enlarge the image with Hi Res Fix
Exercise - End-to-end workflow 1

Quiz – checkpoint models

What is the native resolution of v1 models? *

512×512 pixels

768×768 pixels

1024×1024 pixels

What is the native resolution of SDXL models? *

512×512 pixels

768×768 pixels

1024×1024 pixels

Where can you find models to download? *

Amazon.com

civitai.com

huggingface.co

Youtube.com