[kurtoid]

Silly question: how does OpenAI host/serve it?

[magixx]

I think on professional hardware you can get 80G of memory per GPU and they can likely do memory pooling.