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

Stable Diffusion - Level 1

Generate your first images

Introduction - Generate your first images

What is Stable Diffusion?

Setting up AUTOMATIC1111

Generate your first images

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 images

Text-to-image settings

Introduction - text-to-image settings

Negative prompt

Batch size and batch count

Sampling method

Recap - text-to-image settings

Quiz - text-to-image settings

Prompt building

Introduction - Prompt Building

Additional details

Prompt building process

Exercise - Prompt Building

Basic Inpainting - Fix small blemishes

Introduction - Basic inpainting

A simple inpainting workflow

Inpainting settings explained

Inpainting models

Quiz - Basic inpainting

Checkpoint models

What are checkpoint models?

How are models made?

Installing a checkpoint model

Popular checkpoint models

Stable Diffusion v2 models

Quiz - checkpoint models

SDXL models

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

Fixing common issues

Introduction - Fixing common issues

Duplicate images

Not showing full body

Messed-up fingers

Quiz - Fixing common issues

How Stable Diffusion work 1

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

End-to-end workflow 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 – SDXL models

What is the native resolution of the SDXL model? *

512×512 pixels

768×768 pixels

1024×1024 pixels

2048×2048 pixels

What are the improvements of the SDXL model? *

Generating Legible text

Follow the prompt more closely

Larger image

Faster generation speed

Finer details

What components do the SDXL pipeline made of? *

Convolutional model

Base model

Autoregressive model

Refiner model

Which are good image sizes for the SDXL model? *

2048 x 2048

1024 x 1024

512 x 512

1216 x 832