ℹ️ Info — These limits apply to the API only — never to your store

Everything on this page counts requests made with an API key against https://api.dzbuild.app/v1.

Your storefront, your customers checking out, your dashboard, the mobile app, and every built-in feature of the platform are completely unaffected. They do not consume this quota, they are never throttled by it, and hitting a 429 here has zero effect on your store staying open and taking orders.

If you have never created an API key, none of this page applies to you.

The per-minute limit is enforced per store — all keys belonging to one store share a single budget, so creating more keys never raises the ceiling. It runs on a fixed 60-second window: the counter resets on each wall-clock minute. Because the window is fixed rather than sliding, a burst that straddles a minute boundary can briefly pass more than the ceiling. Handle 429 defensively rather than pacing exactly to the limit.

The API is available on the Enterprise plan only. Keys can only be issued for Enterprise stores, and a key whose store leaves Enterprise — by downgrade or by letting the subscription expire — gets 403 forbidden ("API access requires an active Enterprise plan") on every call. Nothing needs re-issuing: the same key starts working again the moment the store is back on an active Enterprise plan.

This ceiling is sized for real integration work: syncing a catalog, running a mobile app or custom storefront, or driving the store from an AI assistant. Normal usage does not come close. If a genuine workload needs more, ask support for a per-store override rather than working around the limit.

There is no monthly cap: usage is bounded by the per-minute ceiling only, unless support has set a per-store override with an explicit monthly allowance (see Quotas).

A few endpoints do far more work per call than a normal read: they download and re-encode files, or start background jobs. These carry a second, smaller budget per store, on top of the per-minute limit above:

Exceeding the per-minute portion returns 429 rate_limited; exceeding the parallel portion returns 429 too_many_concurrent, which simply means "retry in a few seconds" — an earlier call is still finishing.

These are deliberately tight. Adding a product's full photo gallery still takes well under a minute, and the cap keeps one integration from slowing the servers that also serve your storefront.

The monthly counter increments once per authenticated request that is not answered from cache — cached GETs and /v1/ping are free, and failed requests still count.

Call GET /v1/quotas (see Quotas) to read the limits currently recorded for your store, including any override support has applied to your account.

A per-store override can raise your allowances without changing your plan. If you are building something that genuinely needs sustained higher throughput, contact support with your expected request pattern.