Seed definitely helps but only for one pose. The seed is the random noise pattern that the AI resolves into an image so if you want the same character I. Multiple situations or poses, I havenβt found the seed helpful.
In AI art, a seed is a starting point for the random number generator. The random numbers are used to make black and white noise, which is what gets turned into the image.
So if you use the same seed, the images turn out nearly the same, or exactly the same.
Thanks for such nice post.
Is there any possibility in these models to provide multiple different prompts at once and create bulk images?
Yes. That mostly depends on the tool you are using.
If you are using Auto1111 for example, you can load prompts from a text box or file
https://github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/2332
What are you trying to do?
Very useful! Some people use "seed" in MJ for consistency. Have you found that it improves your results in any way?
Seed definitely helps but only for one pose. The seed is the random noise pattern that the AI resolves into an image so if you want the same character I. Multiple situations or poses, I havenβt found the seed helpful.
What is "seed"?
In AI art, a seed is a starting point for the random number generator. The random numbers are used to make black and white noise, which is what gets turned into the image.
So if you use the same seed, the images turn out nearly the same, or exactly the same.
Here is more detail - https://docs.midjourney.com/docs/seeds
Thanks!