Make responses start faster by removing unnecessary cleanup calls #6625
Conversation
This file contains 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
jfmherokiller
pushed a commit
to jfmherokiller/text-generation-webui
that referenced
this pull request
Jan 15, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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Outdated suggestions cannot be applied.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.
The
clear_torch_cache()function takes about 0.08 seconds to run because it includes a call togc.collect(). Previously, this function was called twice before each generation to address memory leaks in Transformers during text streaming.Changes made:
clear_torch_cache()calls for loaders other than Transformers, saving approximately 0.2 seconds per generation and making replies start faster both in the UI and the API.clear_torch_cache()for Transformers from two to one, cutting the time spent on this function by half.