Skip to content

The method transfer the KV cache from cpu memory to gpu memory #3

Description

@digbangbang

Wonderful work! Following Q and looking forward ur reply.

  1. I am curious about the method in your paper that copy the KV cache from cpu memory to gpu memory.
image

Since I have test the following code

model = nn.Linear()

model.to('cuda')

sometimes the model is large, then the time is costly.

  1. Otherwise, does the method in ur paper is time efficient?

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

      Milestone

      No milestone

      Relationships

      None yet

      Development

      No branches or pull requests

      Issue actions