Preserve mlx-rl checkpoint identity in CISPO settlement - #117
Merged
JoshuaPurtell merged 1 commit intoSep 14, 2026
Conversation
JoshuaPurtell
merged commit Sep 14, 2026
6fc99c5
into
claude/workshop-v011-rc1-20260911
1 check passed
JoshuaPurtell
deleted the
codex/workshop-optimizer-residual-consolidation-20260914
branch
September 14, 2026 07:12
3 tasks
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Carries the one residual behavior from stale PR #89 that was not already superseded by the current internal optimizer architecture: mlx-rl checkpoint items may expose their stable identity as item id without repeating checkpointId or checkpoint_id. The kernel now accepts that identity and proves it settles as the policy checkpoint.
The current internal line already contains the projected optimization_algorithms admission, SFT dataset validation event, and bounded sidecar polling retries from #89, so those old conflicting changes are not replayed.
Validation: focused Rust regression passes (1/1); diff check clean.