Usage and billing

Customers usually buy a tier that unlocks a level of calls, rather than paying purely per invocation. Metered surfaces include server functions, API functions, variables, AI functions, Tabi queries, and similar. Exact invoice line items are on PolyAPI pricing. This page does not republish price tables.

What is not a server execution

A Client Functions runs in-process. It is not billed as its own server function execution. Nested client calls inside one server function still sit inside that one server execution.

Client-only work (no server function) is not billed as executions. The tier still applies to other resources you touch (API functions, Vari, Tabi, AI Functions, …).

When you call a PUBLIC function owned by another tenant, metering is against the caller’s tenant.

AI Function execute has a separate meter from server function executions. See AI Functions.

alwaysOn

alwaysOn is a separate add-on (pricePerAlwaysOnFunction). It keeps a warm pod. It does not change per-execution pricing. See Limits and timeouts.

Rate limits vs invoices

HTTP 429 means you hit a rate or inventory cap (calls per day/minute, maxFunctions, …). That is not the same as an invoice line. See Limits and timeouts.