Same seed, same prompt, same settings, same frame. That reproducibility is what turns generation from a slot machine into an experiment: change exactly one variable and you can see what it did.
It is also the cheapest debugging tool in the stack. When a set stops working, holding the seed and varying one term at a time finds the culprit in minutes. A hundred rerolls finds nothing.
QUESTIONS PEOPLE ASK
What does the seed do in image generation?
It sets the initial noise. Fixing it makes results repeatable, so you can isolate the effect of a prompt change, a guidance change or a model change.