Skip to content

Add Fast KVzip#183

Open
Janghyun1230 wants to merge 3 commits intoNVIDIA:mainfrom
Janghyun1230:fastkvzip
Open

Add Fast KVzip#183
Janghyun1230 wants to merge 3 commits intoNVIDIA:mainfrom
Janghyun1230:fastkvzip

Conversation

@Janghyun1230
Copy link
Contributor

PR description

This PR adds a new KV eviction method, Fast KVzip:
arxiv: https://arxiv.org/abs/2601.17668
code: https://github.com/Janghyun1230/FastKVzip

Checklist

Before submitting a PR, please make sure:

  • Tests are working (make test)

  • Code is formatted correctly (make style, on errors try fix with make format)

  • Copyright header is included

  • All commits are signed-off using git commit -s

  • (new press) mypress_press.py is in the presses directory

  • (new press) MyPress is in __init__.py

  • (new press) README.md is updated with a 1 liner about the new press in the Available presses section

  • (new press) New press is in the default_presses list in tests/default_presses.py

  • (new press) A docstring is provided that follows the same structure as the existing ones

[Note] My code passed tests except for the permission deny regarding the expected attention stats:

FAILED tests/presses/test_ea_with_stats.py::test_load_stats - huggingface_hub...
E               403 Forbidden: You don't have the required permissions to complete this action.
E               Cannot access content at: https://huggingface.co/api/collections/alessiodevoto/expectedattentionstats-68b0248d519303713320e2cf.

Signed-off-by: Jang-Hyun <kimjanghyun1230@gmail.com>
Signed-off-by: Jang-Hyun <kimjanghyun1230@gmail.com>
Signed-off-by: Jang-Hyun <kimjanghyun1230@gmail.com>
@copy-pr-bot
Copy link

copy-pr-bot bot commented Feb 6, 2026

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant