fix(openai): preserve the GPT Live cumulative usage watermark - #7231
Open
kiwijuicer wants to merge 2 commits into
Open
fix(openai): preserve the GPT Live cumulative usage watermark#7231kiwijuicer wants to merge 2 commits into
kiwijuicer wants to merge 2 commits into
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
GPT Live usage is cumulative, but every report currently replaces the watermark on upstream. Reports of 10, 8, and 12 seconds produce deltas of 10, 0, and 4 seconds, overcounting usage. Invalid values can also contaminate later metrics or prevent the final close event from acknowledging shutdown.
Advance the watermark only for finite, nonnegative numeric durations strictly greater than the previous total. Validate raw usage before OpenAI's event constructor coerces values: oversized integers otherwise raise during conversion, and booleans can become numeric durations. Invalid usage is ignored while the enclosing
session.closedevent still completes its acknowledgement. The original event payload is preserved.Regression coverage includes duplicate/regressive reports, malformed and non-finite values, oversized positive and negative integers, boolean coercion, and final-close usage. The GPT Live and duplex adapter suites pass (114 tests), as does
make check.