Two runs, side by side

Gemini 4 Argon · pymd-heading-offset · dumber: Meh → Dumb

Now: Repeated mistake — stuck past the hard cap

Before: Too slow — 1132s over 300s

Claude Opus 5.5 · larder-i18n · smarter: Meh → Smart

Now: language negotiation and every message in Korean

Before: Too slow — 127s over 53s

Claude Opus 5.5 · larder-sales · smarter: Meh → Smart

Now: sales import idempotent and atomic, variance right, docs kept

Before: Too slow — 180s over 140s

Claude Sonnet 5.5 · larder-sales · smarter: Dumb → Meh

Now: Poor output — said done but admitted a gap: failed: tests

Before: Hallucination — claimed success: failed: imports

GPT-6.1 Terra · larder-fefo · smarter: Meh → Smart

Now: FEFO by the data rules, with a regression test

Before: Too slow — 42s over 35s

Gemini 4 Argon · pymd-tasklist · smarter: Dumb → Meh

Now: Too slow — 1136s over 300s

Before: Repeated mistake — stuck past the hard cap