Skip to content

Mixed style of chunked prefill #1013

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
merged 23 commits into from
Aug 16, 2024
Merged

Mixed style of chunked prefill #1013

merged 23 commits into from
Aug 16, 2024

Conversation

hnyls2002
Copy link
Collaborator

Thank you for your contribution, we really appreciate it. The following instructions will help improve your pull request and make it easier to receive feedback. If there are any items you don't understand, don't worry. Just submit the pull request and ask the maintainers for help.

Motivation

This is to reduce the first token latency.

Modification

When mixed style is turned on, try to fill some prefill tokens into the running batch.

Checklist

  1. Ensure pre-commit pre-commit run --all-files or other linting tools are used to fix potential lint issues.
  2. Confirm that modifications are covered by complete unit tests. If not, please add more unit tests for correctness.
  3. Modify documentation as needed, such as docstrings or example tutorials.

@hnyls2002 hnyls2002 marked this pull request as draft August 10, 2024 19:00
@zhyncs zhyncs mentioned this pull request Aug 11, 2024
Merged
@merrymercy merrymercy marked this pull request as ready for review August 13, 2024 09:31
@zhyncs zhyncs added enhancement New feature or request high priority labels Aug 13, 2024
@merrymercy
Copy link
Contributor

Can you resolve the conflicts?

@merrymercy
Copy link
Contributor

  1. Add some unit tests to CI (https://github.com/sgl-project/sglang/blob/main/test/srt/test_eval_accuracy_large.py, https://github.com/sgl-project/sglang/blob/main/test/srt/test_eval_accuracy_large_chunked_prefill.py, https://github.com/sgl-project/sglang/blob/main/test/srt/test_eval_accuracy_mini.py)
  2. Run the full accuracy tests, but do not need to add them to CI
  3. It can then be merged

merrymercy
merrymercy reviewed Aug 16, 2024
View reviewed changes
@merrymercy merrymercy enabled auto-merge (squash) August 16, 2024 08:53
@merrymercy merrymercy merged commit 3694f8f into main Aug 16, 2024
6 checks passed
@merrymercy merrymercy deleted the mix-chunk branch August 16, 2024 09:13
hnyls2002 added a commit to yichuan-w/sglang that referenced this pull request Aug 16, 2024
@hnyls2002
Mixed style of chunked prefill (sgl-project#1013)
bf6cc16
@merrymercy merrymercy mentioned this pull request Aug 27, 2024
Closed
29 tasks
timethink pushed a commit to timethink/sglang that referenced this pull request Mar 9, 2025
@hnyls2002
Mixed style of chunked prefill (sgl-project#1013)
b0bd2d7
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@merrymercy merrymercy merrymercy approved these changes

@Ying1123 Ying1123 Awaiting requested review from Ying1123 Ying1123 is a code owner

@ispobock ispobock Awaiting requested review from ispobock ispobock is a code owner

@zhyncs zhyncs Awaiting requested review from zhyncs zhyncs is a code owner

Assignees
No one assigned
Labels
enhancement New feature or request high priority
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@hnyls2002 @merrymercy @zhyncs