See the call
AI calls, agents, and APIs in your app runtime — tokens, retries, and tool fanout, not just HTTP.
meter ai_tokens=5247
function=OpenAI.post · path=/api/chat
function=OpenAI.post · path=/api/chat
AI usage metering UsageFlow
AI usage metering that blocks overages before they run
Meter AI calls in your runtime and enforce spend before execution — so overages never leave your app.
Free for 10K calls/month · JavaScript, Python & Go
Meter → enforce → cover
One runtime path. Usage becomes a decision before the provider bills you — not a dashboard after the fact.
Hold the switch
Ledger policies decide allow or block before execution. Overages stop in your runtime — not in next month’s invoice.
ENFORCE blocked (478)
reason=tenant_spend_cap · $100.00
reason=tenant_spend_cap · $100.00
Agents and APIs
One platform for AI agents, model calls, and traditional endpoints — function-level where cost matters.
scope agent | api
function-level · path=/api/*
function-level · path=/api/*
Built for teams who need the switch
Gateways see HTTP. UsageFlow meters function paths — then enforces before the provider call runs.
-
Engineers
Install the agent; meter and block in runtime. Skip building a custom metering stack. Pairs with the AI agent meter platform when agents drive spend.
-
FinOps
Enforceable spend caps tied to live traffic — not after-the-fact cost dashboards.
-
Founders
Know what runs, what gets blocked, and what can sync to Stripe when you’re ready to charge.
FAQ
Straight answers for the category questions.
What is an AI meter?
An AI meter tracks AI usage — tokens, calls, and infrastructure — in your app runtime so you can enforce spend before costly work runs and bill accurately when you’re ready.
How does AI usage metering work with UsageFlow?
Install a UsageFlow agent in your runtime. It meters AI calls and APIs, applies Ledger policies, and blocks overages before execution. Optionally, configure a Stripe API key so UsageFlow can fire meter events to your Stripe meters.
Is UsageFlow only for AI agents?
No. UsageFlow meters AI agents, model calls, and traditional API endpoints in one platform — function-level where cost matters, API-level everywhere else.
Meter AI usage. Block overages before they run.
Free for 10K calls/month · JavaScript, Python & Go