Comment by yumraj
16 hours ago
Where are you getting 60GB from? It shouldn’t be that large.
But yes, would love to save context/cache such that it can be played back/referred to if needed.
/compact is a little black box that I just have to trust that is keeping the important bits.
The KV cache consists of activation vectors for every attention head at every layer of the model for every token, so it gets quite large. ChatGPT also estimates 60-100GB for full token context of an Opus-sized model:
https://chatgpt.com/share/69dc5030-268c-83e8-92c2-6cef962dc5...
That is actually nuts.... I'm trying to understand the true costs of AI, wonder how I plug this in!
There are ways to quantize or compress KV cache down.