fix(bench): compressor retry rotates to a named fallback model#256
Merged
Conversation
Same-model retries do not help when the model itself is degenerate-mooding (two empty replies in a row from the primary on the same prompt) — the authorStrategy lesson applied to the compressor: the retry goes to COMPRESSOR_FALLBACK_MODEL (default gpt-4o-mini, probe-verified).
tangletools
approved these changes
Jun 11, 2026
tangletools
left a comment
tangletools
left a comment
Contributor
There was a problem hiding this comment.
✅ Auto-approved PR — 8c6b60d9
Blanket team auto-approval is enabled for this reviewer service.
The full PR reviewer audit still runs separately and will publish findings if it detects issues.
tangletools · auto-approval · reason: blanket_auto_approve · 2026-06-11T00:38:35Z
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
The honest-billing AIME grid re-run died at the κ cell: the compressor returned empty TWICE in a row (primary + same-model retry) — same-model retries don't help when the model itself is having a degenerate episode. The authorStrategy lesson applied: the retry now rotates to
COMPRESSOR_FALLBACK_MODEL(defaultgpt-4o-mini, probe-verified on this prompt). Fail-loud floor unchanged. Re-run relaunches on this.