Yep. I was thinking of something specifically designed for LLMs though. Aka massive memory pools, among other things. Right now, the I think the model has to b

Yep. I was thinking of something specifically designed for LLMs though. Aka massive memory pools, among other things. Right now, the I think the model has to be split up for inference, which is inefficient.