AI engineering glossary · Generative media multimodal
What is guidance scale?
Guidance scale (also called CFG scale): The number that controls how strongly a diffusion model is pushed to follow the prompt during generation. Low values give varied but loosely related pictures; high values follow the text closely but can oversaturate colours and reduce variety.
Guidance scale: an example
Setting the scale to 3 for 'a masala chai stall at dusk' gives soft, creative results, while 12 gives a literal but harsher image.
Related terms
More Generative media multimodal terms
- Image captioning
- Image tokenizer
- Image-to-image translation
- Inception Score
- Inpainting
- Late fusion
- Latent diffusion model
- Latent space
Where you learn it
Guidance scale is part of the AI Engineering Program, which teaches it with notes, a hands-on lab and an exam. The first lesson is free.