Supports vnni-256 for GPTQ INT4#1926
Merged
yyj6666667 merged 4 commits intokvcache-ai:mainfrom Apr 13, 2026
Merged
Conversation
Contributor
|
Warning Gemini is experiencing higher than usual traffic and was unable to create the review. Please try again in a few hours by commenting |
ErvinXie
requested review from
mrhaoxx and
ouqingliang
and removed request for
ouqingliang
Collaborator
|
here is a prefill speed comparison: "PTok" represents the number of prompt tokens processed during the prefill stage. It offers a stable ~5x acceleration in prefill stage, especially as the prompt length increases. |
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
2 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What does this PR do?
Add avx-vnni-256 backend for GPTQ INT4.There is a significant improvement in the prefill.The accuracy loss is around 0.013, which is larger than AVX2's 0.003.