Generative AILesson 5
Lesson 58 min
Diffusion (Image Models)
How AI image generators turn noise into pictures.
What you will learn
- ✓The denoising idea behind diffusion
- ✓How text guides the image
- ✓Why prompts and seeds matter
Explanation
Image models like Stable Diffusion work by denoising. During training they learn to remove noise from images step by step.
To generate, they start from pure random noise and iteratively clean it up, guided by your text prompt, until a coherent image emerges.
The text prompt steers the direction, and a random seed sets the starting noise — same prompt + same seed gives the same image, which is handy for reproducibility.
Real-world use
When you tweak a prompt and the image changes dramatically, you're steering that denoising process toward a different result.
Common mistakes
- • Expecting identical images from the same prompt — the random seed changes the output unless you fix it.
Practice
Write the same image prompt two ways (vague vs detailed) and predict which gives a more controlled result.
Knowledge check
0/1 answered1. Diffusion image models generate by...
Answer all questions to check.