vectorizers(litellm): Support setting max chunks per batch using env var #898
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
|
|
Contributor
Author
|
An alternative, config-less implementation you could consider of exponential increase: start with a small size, say 4, and on each request, double that, until you hit an error, then back off to the last successful batch size. Sloppier, and more complicated to implement initially, but then should require much less tuning. |
Collaborator
|
Hey @NickCrews thanks for the PR. I'll make some time to review. Can you please rebase against main. I was fixing some things for pg18. |
Contributor
Author
|
@alejandrodnm I think I fixed the lint error and the format of the commit message |
7cca0f0 to
e67ccc2
Compare
- max chunks per batch when sending using "gemini" - add ability to configure this even more arbitrarily using env vars Signed-off-by: Nick Crews <nicholas.b.crews@gmail.com>
e67ccc2 to
5137c03
Compare
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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 am using gemini as my provider. This wasn't listed as one of the providers, so I found I was getting the default of 5, and a warning message. This was slow.
So I added gemini to the code so it was supported.
But, more in general, I don't want some other user to be forced to modify the code in order to support their provider, so now you can configure this param with even more control using an env var