fix: reranking models limited to 512 tokens in llama.cpp backend #6344
Conversation
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
✅ Deploy Preview for localai ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
…ackend Signed-off-by: JonGames <18472148+jongames@users.noreply.github.com>
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Outdated suggestions cannot be applied.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.
Description
I discovered a bug with how llamacpp's grcp backend is implemented in LocalAI. Basically the batch parameter is exposed in the LocalAI model yaml file but this only changes the n_batch parameter in llama.cpp. When trying to run a reranking model in llama.cpp, it sometimes complains that the prompt input is larger than the maximum physical batch size. The default is 512 tokens and set by n_ubatch, not by n_batch. I fixed it by making the batch value from the model yaml file change both the n_batch and n_ubatch parameters in llamacpp.
Notes for Reviewers
Signed commits