Jason Antic

1/ With DeOldify being a real business now we invested in 4 Quadro RTX 8000 48GB video cards back in March. Expensive- definitely. Worth it? Totally. I'm now having the revelation of "oh...bigger models and bigger batches do help!" rather than endlessly chasing clever hacks.

I am little bit confused. That means are you continuously training to increase it's performence? I thought when a model is trained, you don't need so much comupting power. Or it is because everyone is trying different pictures parallelly.

They don't get continuously trained once completed. You just train a new model until completion and then use it without further training. So I'm talking about making new models here that I couldn't previously make because of new hardware.

OK that makes sense. Thanks for the reply. So from the Tweet I just understood your first setup was 4 1080Ti, with what you implemented your first deoldify.