Record every LLM call in the trace - #68
Merged
Merged
Conversation
A call that returned no assistant text emitted no span, so a turn whose last call came back empty read as ending on a tool call. Those are the calls worth seeing -- they end turns and still cost tokens. Only the context events stay on the first call; they repeat verbatim.
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
A call that returned no assistant text emitted no span:
So a turn read as ending on a tool call, with no LLM call after it — which is impossible, and is what made the reported trace look wrong. Those are exactly the calls worth seeing: they end turns and still cost tokens (
usage_output=4per attempt in the retry storm).The stated intent was avoiding duplicated context events across a tool-call loop. That only requires dropping the events, not the span.