Budgets for paid API calls
Set spending limits per request and per session for API, inference, and compute usage.
Limits for API spending
Four rules stop runaway spend: per-call, session, provider, and velocity. A cheap call from a listed provider clears with no review step.
Per-call ceilings
A single API or inference request can't cost more than the cap you set for the task.
Session budgets
Give each task or agent run a hard budget so many small calls cannot add up to an unlimited bill.
Approved providers
Allow paid requests only to the providers or service types you choose.
Velocity monitoring
Limit repeated paid requests and retries within a set period.
No single call escapes the session budget
A buyer agent shops for inference call by call under a $5 per-call cap and a $25 session budget. A $9 call is refused, and so is the call that would push the session past $25.
Enchant runs on Conto budgets
Enchant pays for tools, data, and model calls per request over x402 and MPP, two pay-per-call protocols. Every user gets a Conto budget, and every paid step is checked against it first.
Visit Enchant at askenchant.comUser asks Enchant
A task can require a paid tool, data source, or model call.
Price appears first
The paid step shows its cost before anything runs.
Conto checks policy
Budgets, per-call caps, and approvals are evaluated before spend.
x402 or MPP clears
Approved micropayments settle and update the user budget.
x402 and MPP micropayments
Conto budgets for every user
Policy before each paid request clears