Articles on: Images

What is a Seed?

The “seed“ parameter in stable diffusion is a number used to initialize the generation process. The seed parameter can be set to a specific value or left to be randomly generated by the model. By controlling the seed parameter, it is possible to generate reproducible images, experiment with different model parameters, or create variations of a prompt. Using the same seed value and prompt can produce the same image, while changing the seed or prompt can result in different outputs. The text provides examples of using the seed parameter to modify the appearance of a generated image by changing the prompt or to create different variations of the same image by using the same seed value.

A good metaphor for the seed parameter in stable diffusion could be the "ingredients" in a recipe. Just like a recipe requires a specific combination of ingredients to produce a particular dish, the seed parameter in stable diffusion requires a specific value to generate a particular image. Changing the value of the seed parameter is like changing the ingredients in a recipe, which can result in different variations of the final product.

Just as a recipe can be reproduced by using the same ingredients, stable diffusion can produce reproducible images by using the same seed value and prompt.

To set your seed value in Scenario, you will need to make sure you have your image output number set to 1. Then you can indicate what you want the fixed seed number to be. This can be a useful way to compare guidance levels, adjust colors, and more while still keeping the base composition nearly the same.

You can find an images seed by clicking on the individual image.

Updated on: 18/01/2023

Was this article helpful?

Share your feedback


Thank you!