Measured on every call
The real cost of every call is recorded with the draft it served. The console shows it per conversation and per month.
Every model call is measured and visible in the console, per conversation. A monthly cap per space, checked before each call, bounds the bill before it exists.
An AI bill shows up at the end of the month, and nobody knows which conversation ran it up. A spike in volume becomes a spike in spend with no warning.
The real cost of every call is recorded with the draft it served. The console shows it per conversation and per month.
Before every paid call, the month’s cost is compared with the plan’s cap. At the cap, the call is not made, the space pauses until the next month and administrators are notified.
Usage is billed back in arrears at your plan’s coefficient, on the same measurement you see.

On the space’s costs page, call by call, and in the drafts log.
No more model calls until the first day of the next month. Your agents keep answering normally; administrators are notified on the first refusal.
It is the measurement the invoice is computed from, multiplied by your plan’s coefficient. The pricing page gives the coefficient and the cap.
Connect an inbox, approve the suggested knowledge base, and get your first prepared replies.
Off, draft or autopilot per inbox; autopilot opens per category once drafts are measured. Refused outright on payments inboxes and unofficial WhatsApp.
Every night, 0seat extracts question-and-answer pairs from your resolved conversations. Nothing is used until an agent has approved it.
The draft arrives as a private note with the knowledge base entries behind it. The agent sends, edits or discards it. Below the threshold, no reply.