CODING · EDITORIAL CLASSIFICATION · NOT A BENCHMARK
DO I NEED THE BIG MODEL FOR…
Review a pull request for style issues
Small-model jobNO — START SMALL
“Save the strong model for correctness; style is mostly policy matching.”
WHY THIS VERDICTSTART WITH
START WITH
THE WORK.
- 01
The work is constrained enough to evaluate cheaply.
- 02
A clear schema, rubric, or example carries much of the workload.
- 03
Most failures can be detected before they become expensive.
RECOMMENDED SETUPSmall, fast model + schema + deterministic checks
NON-NEGOTIABLE
Back the review with formatter, lint, and repository conventions.
REASONING2/5
AMBIGUITY2/5
ERROR COST1/5
CONTEXT3/5