From the course: Stable Diffusion: Tips, Tricks, and Techniques
Unlock the full course today
Join today to access over 24,600 courses taught by industry experts.
Writing prompts for series 2 models - Stable Diffusion Tutorial
From the course: Stable Diffusion: Tips, Tricks, and Techniques
Writing prompts for series 2 models
- [Instructor] In Stable Diffusion terms, a model is simply the set of trained data that serves as the basis for all imaging. As I mentioned earlier Stable Diffusion models fall into two categories, version 1.X models and version 2.X models. When you train a new model for stable diffusion you use an existing model as the basis for the one you're training. If you find a new model to download, something we will look at later, you should see a listing for what source model, or in this case base model, that this model was generated from. And this will usually be 1.4, 1.5 or 2.1. Everything that we have looked at so far, in terms of prompt writing, applies to any model that was based on the original 1.4 or 1.5 version. Version 2.1 has important improvements over version 2.0, so most people who want to make a version derived from the 2.0 series use the 2.1 model as a source. Now when I say a version 2.0 model what I really…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
What does a prompt do?3m 37s
-
(Locked)
Stable Diffusion seeds2m 1s
-
(Locked)
Stable Diffusion batches and pixel counts3m 42s
-
(Locked)
Prompt basics11m 6s
-
(Locked)
Questions to answer when writing prompts5m 36s
-
(Locked)
PNG information and saving8m 32s
-
(Locked)
Using CFG scale6m 1s
-
(Locked)
Prompt weighting5m 23s
-
(Locked)
Writing prompts for series 2 models6m 48s
-
(Locked)
Prompt libraries and styles4m 22s
-
(Locked)
Interrogating an image2m 47s
-
(Locked)
Artist names and rendering styles3m 57s
-
-
-
-
-
-
-