Repository navigation
WIP: Judge tuning (3/6): Preserve native usage in caches and annotations - #140
Draft
ErlisLushtaku wants to merge 11 commits into
Draft
ErlisLushtaku wants to merge 11 commits into
ErlisLushtaku wants to merge 11 commits into
Conversation
…lislushtaku/feat/cache-native-usage
This was referenced Oct 9, 2026
…lislushtaku/feat/cache-native-usage
…lislushtaku/feat/cache-native-usage
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
WIP. Not ready for review yet.
Preserves native RequestUsage alongside completion and judgement cache outputs and carries it into meta-eval annotations. Cache hits restore token counts without adding a new request or charge to the current run's usage tracker.
Existing SQLite databases gain a nullable usage column, preserving their rows. Cache sharing accepts older incoming databases without changing them. Skipped context-limit requests remain uncached failures.
This is the prerequisite for native-token cost objectives in #133. It is stacked on #135 and includes the existing #138 context-failure prerequisite; merge #138 before this PR so its independent review is preserved.
Part 3 of 6. Merge order is #137, #134, #135, this PR, #133, #139, then #136, with #138 merged before this PR.