---
title: "Limits"
description: "Rate limits and quotas per app and per visitor, what files cost, and what happens past a limit."
---

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

# Limits

## Per app

Calls are counted per app, in each store: every token an app holds counts against that app's limit, whichever customer it's for. An app makes 600 calls a minute unless the merchant set it otherwise for the app, under **Settings**, **Connections**, **API clients**. The merchant may also give it a quota per day or per month, counted in UTC.

## Per visitor

A publishable key's rate limit holds per visitor instead: each address calling with it (an IPv6 address by its /64) gets the key's calls a minute, so one visitor busy on your site never holds back the others, though visitors behind one shared address share it. Recording a page a visitor saw ([Visits](/customer-api/storefront#visits)) counts too. A quota set on a key caps every visitor's calls together, and since the key is public, anyone holding it can spend it.

## Files

A private file costs two calls each time: asking for its URL, and fetching it. Asking again once the URL's `expires_at` passed costs two more, though a video asked for with a publishable key holds 4 hours. Each range fetched is a fetch, ranges a video element asks for on its own included, though fetches of one URL from one address within a few seconds of the first may share its call. See [Files and media](/customer-api/files).

## Past a limit

A call past a limit answers `429` [`not_permitted`](/customer-api/errors#not_permitted) with the reason `limit`, and a `Retry-After` header when the service knows when to retry. Wait that long, and slow down.

## Refused credentials

Tokens a store refuses (unknown, expired, revoked or malformed) are also limited per network. Past about 60 a minute from one address (an IPv6 address by its /64), or 600 across an IPv6 /48, calls to that store with a token the store hasn't accepted in the last few minutes answer `429` with the reason `limit` for up to a minute, never `invalid_token`. Keep the token and retry after `Retry-After`. A token in steady use keeps working.

Publishable keys a store refuses count the same way, so a page whose key was revoked answers `429` for a while past that many calls, while a key in use keeps working.

## Email codes

Codes that prove a new address are limited per hour, a code sent again included, whatever app or page asked: 5 for one customer, 3 of them to one address, 20 to one address across customers, and 200 for the store across its customers. A customer may also ask for 5 address changes an hour from one network. Each day (in UTC), one customer may be sent 10 and one address 20. Past any of these, [`POST /contacts/change`](/customer-api/reference/change-contact) answers `429` with the reason `limit` until the hour or the day ends.

## Subscriptions

A customer may cancel or resume one subscription 6 times in any 24 hours. Past that, [`POST /subscriptions/{subscription}/cancel`](/customer-api/reference/cancel-subscription) and [`POST /subscriptions/{subscription}/resume`](/customer-api/reference/resume-subscription) answer `429` with the reason `limit`, and `Retry-After` says when the oldest of those frees. The store's emails about them wait 2 minutes, so a change taken back at once sends only the latest.

## Browsers

A call carrying an `Origin` header costs nothing more: reading which origins may read the answer isn't counted. A preflight is kept 10 minutes per URL, so a page reading many products pays one preflight per product URL.

## Bodies

A request body holds at most 16 MiB (`413` past it), nests at most 64 levels deep, and holds at most 100000 objects and arrays and 100000 values (`422` past them).

Source: https://docs.sherpo.io/customer-api/limits/index.mdx
