Skip to content

Conversation

ispobock
Copy link
Collaborator

@ispobock ispobock commented Feb 18, 2025

Motivation

For draft decode in spec, the max batch size should be req_to_token_pool.size*topk.

@ispobock
Copy link
Collaborator Author

cc: @Ying1123

@zhyncs zhyncs merged commit e5ce395 into sgl-project:main Feb 18, 2025
18 of 20 checks passed
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