Skip to content

Support stop_token_ids in sglang API#1092

Merged
hnyls2002 merged 6 commits intomainfrom
stop-token-ids
Aug 15, 2024
Merged

Support stop_token_ids in sglang API#1092
hnyls2002 merged 6 commits intomainfrom
stop-token-ids

Conversation

@hnyls2002
Copy link
Copy Markdown
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

Support stop_token_ids in SGLang API.

Modification

  1. add API support.
  2. always check whether the current token is in stop_token_ids.

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 requested a review from merrymercy August 14, 2024 08:42
@merrymercy
Copy link
Copy Markdown
Contributor

Looks good. Please fix the CI.

@hnyls2002
Copy link
Copy Markdown
Collaborator Author

@zhyncs Even with retrying, the moe throughput varies a lot. I modified this to 930.

@hnyls2002 hnyls2002 enabled auto-merge (squash) August 15, 2024 00:25
@hnyls2002 hnyls2002 merged commit 73cf683 into main Aug 15, 2024
@hnyls2002 hnyls2002 deleted the stop-token-ids branch August 15, 2024 00:31
timethink pushed a commit to timethink/sglang that referenced this pull request Mar 9, 2025
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.

2 participants