Repository navigation
Fix admitted model readiness with live context limits - #155
Merged
Merged
Conversation
This was referenced Oct 1, 2026
This was referenced Oct 1, 2026
Contributor
Author
|
Independent adversarial review (Claude Opus, run by the operator for Anders, 2 October; different model family from the author). |
This was referenced Oct 2, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Runtime can admit a verified model package while Marketplace Use and cold Init report its service unavailable. The live native provider includes local context limits, and Runtime previously compared those extra fields with a saved activation summary that omitted them.
This change accepts the exact 4096-token, one-slot startup profile and 1024-token output limit. Saved activation bytes, offer identity and policy stay exact; older summaries remain compatible. Malformed or changed limits and extra authority fields fail the comparison.
The fixture-only follow-up aligns explicit native tests with current Keep intent and chat input-count contracts.
Validation: 34 focused regular startup-binding regressions passed; all four explicit native cases have passing evidence across justified runs; the four basic gates passed. Two independent source/security review rounds passed on production diff
8a376cd4…and final fixture diffae6c63e3…. Initial fixture failures remain recorded with their causes in #154.Refs #154. Installed Marketplace Get → Use → visible Assistant reply, final-candidate cold Init, completed CI and required Irzhy review remain open. This draft targets develop; #141 confinement and #98 resource qualification remain separate tasks.