[BFCL] Handling rate limits #559
Merged
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
ShishirPatil
added
enhancement
HuanzhiMao
requested changes
Jul 29, 2024
There was a problem hiding this comment.
retry_count < retry_limit - 1should be removed, otherwise it would only retry two times (whenretry_countis 0 and 1).- You can also check the status code for indication of rate limit reached. Such as
429 Too Many Requests,403 Forbiddenor503 Service Unavailable. Purely relying on the text "rate limit reached" might be inaccurate. For example, the below error message from Anthropic API would not be caught:
Number of requests has exceeded your rate limit. Please try again later or contact sales to discuss your options for a rate limit increase.
CharlieJCJ
requested changes
Aug 4, 2024
HuanzhiMao
approved these changes
Aug 4, 2024
CharlieJCJ
approved these changes
Aug 4, 2024
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Outdated suggestions cannot be applied.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.
For BFCL generations of hosted models, this will check for rate limit errors and retry thrice. We can choose to include an exponential back-off as is usually recommended, but I think in the interest of keeping it simple this should perhaps be sufficient?
This does NOT affect the leaderboard values.