Cost cap
A hard spending limit enforced by your code, per user, per day or per month.
Public AI features can be abused or loop by mistake. A cost cap reserves an estimate before each call, settles the real cost after, and refuses work when the budget is spent. It needs one consistent counter, not a cache that is eventually consistent.
Related terms
- Tokens: The pieces of text models read and write, and the unit you pay for.
- Agentic loop: The cycle of model call, tool call and result that an agent repeats until it stops.
- Guardrails: Checks around a model that block, fix or flag inputs and outputs that break the rules.
