[warkdarrior]

Not sure what you mean by "deterministically", but most models are deterministic during inference: given the same input or prompt, they'll generate the same output.

[pfortuny]

Are they? I thought there was always a kind of “random parameter” or random step midway?

Hope this question is not too dumb!

[ihateolives]

Usually not, perhaps noise filters, but mostly not. There was amusing case of reversing twirl filter a while ago: https://boingboing.net/2007/10/08/untwirling-photo-of.html

[msk-lywenn]

Surely that twirl effect isn't made using AI but just a geometric function, right?

[pfortuny]

Thanks a lot. I was mistaken.

[cfn]

It depends on the algorithm, but many models can be re-generated if you start with the same seed.

[msk-lywenn]

Is it common and/or expected to have control over the seed though?

[comboy]

Yes, if you are fine tuning some other parameters you need to keep the seed the same to be able to have any idea what you're doing.

[pfortuny]

Thank you! I was mistaken then.