fix(eval): smoke_eval resolves models via tiers + sonnet-5 baseline#94
Merged
Conversation
…corded smoke_eval.py hardcoded sonnet-4-6/opus-4-6 — the rag-gate ATTUNE_MODEL_PREMIUM pin never reached it. Models now resolve via attune_author.model_tiers (answers=capable, judge=premium). Two live Claude 5 fixes: no temperature param (400 'deprecated for this model'), first-TEXT-block extraction (responses can lead with a ThinkingBlock). BASELINES.md records the first sonnet-5-judged run: faithfulness 100% (PASS), strict accuracy 60% advisory (was 40%). specs/fable-model-tiers task 10 (attune workspace repo). Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Closes the gap between the rag-gate CI pin and the eval itself:
smoke_eval.pyhardcoded sonnet-4-6 answers / opus-4-6 judge, so the task-8ATTUNE_MODEL_PREMIUM=claude-sonnet-5pin was decorative. Models now resolve viaattune_author.model_tiers(answers = capable tier, judge = premium tier).Two live Claude 5 API fixes from the first sonnet-5 run:
temperatureis rejected outright (400 'deprecated for this model'), and responses can lead with a ThinkingBlock — extract the first TEXT block.First sonnet-5-judged baseline (local, 2026-07-10), recorded in BASELINES.md: Faithfulness 100% (hard gate PASS), Strict Accuracy 60% advisory (up from 40% under the old models); q27/q39 remain the partial stragglers.
specs/fable-model-tiers task 10 + specs/rag-gate-accuracy-baseline (attune workspace repo).
🤖 Generated with Claude Code