Repository navigation

docs(evals): T1 campaign results — LCD vs bare agent (single-shot) - #13

Merged

stid merged 3 commits into

mainfrom

docs/t1-campaign-results

Aug 29, 2026

Commits on Aug 23, 2026
- docs(evals): T1 campaign — 6 rows + conclusion: bare agent ~4.5x cheaper single-shot, quality near-parity
  
  stid
  and
  claude
  committed
  12a3301
Commits on Aug 29, 2026
- docs(evals): verified-isolation rerun — old rows invalidated, clean rows + rewritten conclusion
  
  stid
  and
  claude
  committed
  f53a5fc
- docs(evals): correct wall-clock ratio to r2 data (~5.1x, was stale ~2.5x) — evaluator round-2 finding
  
  stid
  and
  claude
  committed
  5f2ae56