Getting started

Limits

Each plan has a few limits. Your own numbers are on the Billing page and in the Account endpoint.

Credits

On the Free plan, requests stop with 429 quota_exceeded when the month's free credits are used, until they renew. On Pay as you go and monthly plans they keep working, and credits beyond the included ones are billed per 1,000, unless you set a monthly spend limit: then they stop with 429 spend_limit_reached.

Concurrent requests

How many requests can run at the same time. Extra requests get 429 concurrency_limit_reached with a Retry-After header; retry them a moment later.

Many searches at once

Send them with a pingback_url instead: they wait in a queue rather than being refused, and run up to your plan's concurrency at a time (see Async tasks). Up to 10,000 of your searches can wait at once; more get 429 queue_full.

Requests per hour

A safety limit per clock hour on the Free plan. When it is reached the answer is 429 hourly_limit_reached, and Retry-After says how many seconds to wait. Paid plans have no hourly limit.

Reading your limits

GET /v1/account returns concurrency and requests_per_hour, where 0 means no limit, and credits_limit, where null means your credits have no cap.