Repository navigation
docs(evals): T1 campaign results — LCD vs bare agent (single-shot) - #13
Merged
Merged
Conversation
…per single-shot, quality near-parity Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VRajFkyg9aBwbUFMhVaxNe
stid
marked this pull request as draft
August 23, 2026 01:41
Owner
Author
|
HOLD — do not merge. The independent closeout evaluator refuted the campaign's isolation claim: |
…ows + rewritten conclusion Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VRajFkyg9aBwbUFMhVaxNe
stid
marked this pull request as ready for review
August 29, 2026 01:11
….5x) — evaluator round-2 finding Co-Authored-By: Claude Fable 5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01VRajFkyg9aBwbUFMhVaxNe
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Campaign rows + conclusion for the first baseline-arm run (v0.17.0 harness). 3 runs/arm, claude-fable-5, settings-only isolation (config-dir isolation broke auth on the host — noted in the conclusion).
Headline (single-shot, this task class): bare agent ~4.5× cheaper / ~5.8× fewer tokens, all 6 runs green, rubric quality near-parity (lcd 69/75, none 72/75). LCD's multi-session claims remain unmeasured — follow-up longitudinal benchmark (T1b) proposed in the dev harness.
Published per the evidence-in-PR policy; unfavorable direction included by design (the harness is an instrument, not a marketing device).
🤖 Generated with Claude Code
https://claude.ai/code/session_01VRajFkyg9aBwbUFMhVaxNe