feat(presets-parity): assert contextWindows against /model/info; align presets (OPS-732)#43
Merged
Merged
Conversation
… align presets to /model/info (OPS-732) The gateway's /v1/model/info is authoritative for all 41 catalog models since unsigned-paas#1449 (chart 0.4.13, live-verified 41/41 enriched). Preset windows aligned to it in pi + omp (17 corrections each — incl. fixing the qwen3.6-35b 262144 guess to the real 131072, GLM-5.2 lanes to 1M/199984, Kimi-K2.6 to 262144, DeepSeek V3/V4 to real windows). The live parity test now fails on drift (with a >=30-id coverage floor so a metadata collapse can't make it vacuous). pi models.json reformatted by the alignment script (json round-trip); content-identical otherwise.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
/v1/model/infois now authoritative (unsigned-paas#1449, chart 0.4.13 deployed + live-verified 41/41 enriched) — the weekly live parity run now asserts preset contextWindows equalmax_input_tokens(was warn-only), with a ≥30-id coverage floor.pi/models.jsonreformatted by the alignment round-trip.Closes OPS-732 (both halves).
Test plan
bun test3 pass/2 skip;PARITY_LIVE=15/5 green with the assertion active (zero drift against the live gateway);harness-pi:test+harness-omp:testgreen.Need help on this PR? Tag
/codesmithwith what you need. Autofix is disabled.