How much Vram needed for the full context length?

#31
by Aly87 - opened

with no quants.

I wonder how do you run this there 2.63M downloads where do you get the gpu's?

Runpod or other cloud providers?

Runpod A100 80GB is 2USD/hr
So that makes using directly chinese providers cheaper.

What do you mean directly Chinese providers? Through their chat interface? Or API or what? I want to fine-tune the model and “own” the weights not just chat with it

I mean uzing AliPay to pay them directly for API

Sign up or log in to comment