update new template to support weight cache and batch - #36
Merged
Conversation
HaibaraAiChan
approved these changes
Nov 30, 2025
JiuChen0
referenced
this pull request
in JiuChen0/BloomBee
Mar 22, 2026
Co-authored-by: TomekWei <twei11@illinois.edu>
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
I have refactored the original template based on the current main branch, and it now supports some of the system features that XU and Charlie previously added or modified, such as multi-batch size and caching. I tested the LLaMA model and the outputs are correct. With a batch size of 4, the throughput is 7.01 tokens/sec/sequence, and the effective throughput is 27.18 tokens/sec. I’m not sure whether this matches Charlie’s results exactly but the muti-batch seems works correctly.