kvcache reuse support - #1
Open
lishixlnx wants to merge 6 commits into
Open
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
kvcache reuse function support ( only test for Qwen3.5-35B)
currently, client API is not changed and always send full prompt ( Q1+A1+Q2)
amdgpu-server find the Q2 and only send it to model by OGA API appendTokens();
it supports web UI of lemonade;
it supports first prompt with pic, and 2nd prompt without pic; ( OGA's limitation)
also fix bug: first prompt with pic, 2nd prompt with pic without kvcache reuse ( actually, this is a bug in this repo and has nothing to do with kvcache reuse)
still debugging to support copilot client.