Depth is the chip beside the composer. It sets two dials for the discussion in front of you — how many seats take part, and how many rounds they get — so the cost of a turn is something you choose rather than something you find out afterwards.
- Quick — one model, one pass. For a question you would not ask twice.
- Compare — two models, one round. A second opinion.
- Consensus — the whole room, reviewing until they agree or run out. The default.
- Deep — the whole room, twice as long to disagree. For work you cannot redo.
Auto
On the desktop app there is a switch that lets each message pick its own depth. It reads the question, then escalates from what actually happened rather than from how the question looked. A one-line question that turns out to make a command fail gets reviewers added mid-run; a question that reads hard but resolves in one pass does not pay for four seats. A guess about difficulty is a guess; a failing command is a fact.
Keeping the cost down
- A session ceiling stops a discussion when it has spent what you allowed. Set it in Settings, beside the rounds picker.
- Reviewers run on the cheaper tier of their model by default; the lead moves to its cheaper model for small asks, or when a ceiling is three-quarters spent.
- Switching a seat off is the bluntest and most effective control.
See Plans, credit and what a month costs for what the money side looks like.