When should expensive models plan and supervise work executed by cheaper models?
Find the boundary between reasoning that needs a frontier model and execution that can safely move to less expensive workers.
Starting point
How far does the planner-worker split hold? Which parts of planning actually require an expensive model, and which parts can be handled by a cheaper one?
What to track
- The kinds of ambiguity that require frontier reasoning.
- Whether supervision can be triggered only at uncertainty or failure boundaries.
- How plan quality changes execution cost and review burden.
Living record
This page is the durable home for experiments that change where Yesod draws the boundary between planner, supervisor, and worker.