ExllamaV2 optimizations #88
Merged
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
…eneck), further 2x speedup
# Conflicts: # lmformatenforcer/tokenizerprefixtree.py
Merge recent changes
|
Merged, thanks @bdashore3 and @turboderp for the contribution! |
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Outdated suggestions cannot be applied.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.
Currently, building the initial token tree is inefficient and can cause slow ingestion of tokens (for example, a JSON schema). This is evident when using models with large vocab sizes such as cohere command-r, gemma, and qwen. Generation locks up and takes hours to process. These commits help optimize that initial building when creating an ExllamaV2 LMFE filter.
Tests: Run command-r with a JSON schema in TabbyAPI using LMFE v0.9.5, would not start generating. With these commits, generation immediately starts.
References #75
Thanks @turboderp for creating these commits.