Repository navigation
feat(WO73/C3): decode-latency budget gate + bench tool - #339
Merged
Merged
Conversation
BudgetGate now enforces perf.decode_ms_per_row: a declared budget with
no measurement refuses ('never assumed'); a measured value over budget
fails the gate. tools/bench_decode.py produces the measurement — it
resolves the artifact through the index (sha256-verified) and decodes
the model's frozen fixtures in the shipped single-row greedy mode.
First measured budgets land from the CI conformance job; no numbers
are declared here — none exist yet, and none are invented.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Completes C3's second half:
perf.decode_ms_per_rowin an index entry is now enforceable — BudgetGate refuses a declared budget with no supplied measurement and fails an over-budget measurement.tools/bench_decode.pyis the measurement half: index-resolved (sha256-verified) artifact, frozen fixtures, shipped single-row greedy decode, median/max ms per row over N repeats.Deliberately no budget numbers declared yet — the first ones come from running the bench against shipped artifacts (CI conformance job is the natural home), never from assumption.