A subset of Image and Video Generation Models support Seeds. There are two main ways that Image and Video AI models use seeds.
What is a Seed?
Model generates seed value:
Image Models
Stable Diffusion
Stable Diffusion 3.5 Large
Bria 2.3
Adobe Firefly V4 Standard
In this case, each image generated by the model comes with a seed number generated. You can find out more about how to locate and use seeds in this way in this article
User generates seed value:
Image Models
Google Imagen 3
Getty Images
Video Models
Runway Gen-3 Turbo
Runway Gen-4 Turbo
Google Veo2
Stable Video
In this case the model will never generate a seed value for you. If you want to make use of seeds, you will need to populate the ‘Seed’ box for a generated image or video that you like. You can then populate this same value into future generations to make use of the seed.
See the following articles for more information on Seeds and their uses within the creative process: