Skip to content

feat: up batch size - #63

Merged
stephantul merged 3 commits into
mainfrom
no-encode-chunking
Oct 27, 2026
Merged

stephantul merged 3 commits into
mainfrom
no-encode-chunking

Conversation

@stephantul

Copy link
Copy Markdown
Contributor

This PR ups the batch size and contains an encode_flat function which, on average, is faster for users who require a contiguous block of vectors anyway.

@stephantul stephantul changed the title No encode chunking feat: up batch size Oct 27, 2026
@greptile-apps

greptile-apps Bot commented Oct 27, 2026 •

Copy link
Copy Markdown

RetriggerView in GreptileConfidence Score: 4/5

[Medium impact] The PR changes how callers encode sentence batches through the model API.

The PR appears safe to merge, with a non-blocking fix recommended for zero-dimensional models.

Reviews (1) · Last reviewed commit: "feat: add flat encoding, up batch size" · Reviewed by Greptile

Comment thread src/model.rs Resolved
@stephantul
stephantul merged commit 577b577 into main Oct 27, 2026
5 checks passed
@stephantul
stephantul deleted the no-encode-chunking branch October 27, 2026 15:11
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.

1 participant