> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tracks.alpic.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Billing

> Your prepaid balance, what runs cost and what is free

Tracks is prepaid. Your organization has a balance in US dollars, and what its runs consume is taken from it as they go. There is no subscription.

## Your balance

Open your profile menu at the bottom of the sidebar, then **Billing**. It shows your balance and every transaction: top-ups, free credit, what each run cost, and refunds.

* Your first sign-in comes with **\$10 of free credit**, unless you joined through an invitation. Free credit never expires.
* **Add funds** pays by card on Stripe, then brings you back. A top-up is at least $10; your first one is at most $500. Tax is added at payment, and each top-up has an invoice in **Transactions**.

A balance belongs to the organization: every member spends from it. See [organization and sharing](https://docs.tracks.alpic.ai/concepts/organization.md).

## What is charged

Each run pays for what it consumes, at the provider's list price:

* **The agent's model requests**. For Cursor, what Cursor charged.
* **Sandbox time**: how long the run's sandbox was up, priced by its size.
* **Replies to the agent's questions**: in an experiment, when an agent stops to ask something, a model writes the user's answer. That call is billed with the run.

`tracks runs usage <id>` and the run's page in the web app show what a run cost.

## What is free

* **Judging**: evaluating each run against its scenario's evaluations.
* **Agent tool use**: counting how often a run used each of your skills, MCP servers and command-line tools.
* **Importing** scenarios or evaluations from a document.

None of these count against a run's budget.

## Limits that protect your balance

**Budget per run.** Every run has a budget: once its cost reaches it, the run ends as `budget_exceeded`. It can go slightly over by the requests already in flight. Defaults and ranges are in [budget and time limit](https://docs.tracks.alpic.ai/concepts/runs.md#budget-and-time-limit).

**Maximum spend.** Before launching an experiment, the **Launch** dialog shows how many runs start and the most they can spend: the number of runs times their budget. It warns you when that is above your balance.

\*\*Balance at $0.** No run starts while your balance is $0 or below, and reaching \$0 cancels the runs still running. The API refuses new runs with a link to add funds.

> [!TIP]
>
> Start with a small budget and one repetition. Raise them once you know what a run of your task costs.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.