DocumentationAPI v1
Documentation/Guides/Rate limits
GuideFree

Rate limits

Two limits, and every response says where you stand on both.

Each plan has a rate for the minute and an allowance for the day (free key) or the month (paid plans). Every response carries both in its headers, so a client can pace itself and never be refused.

x-ratelimit-limit: 60
x-ratelimit-remaining: 58
x-ratelimit-reset: 41          seconds until the minute resets
x-allowance-limit: 500
x-allowance-remaining: 487
x-allowance-reset: 34201       seconds until the day resets

Going over

A call past the minute answers 429 rate_limited; a call once the allowance is used answers 429 allowance_used. Both carry a Retry-After header in seconds and a sentence saying when to come back. Allowances renew at 00:00 UTC. There is no overage bill, ever: a refused call is simply refused.

What counts

Only calls that are authorised and understood count. A call refused for its key, its plan or its parameters costs nothing.

PlanEndpointsWindowRateAllowance
Free key16Every free endpoint, on results at least seven days old60 a minute500 a day
Live44In-play and the last thirty days120 a minute250,000 a month
Archive43The whole archive, but nothing in play120 a minute250,000 a month
Complete44In-play and the whole archive180 a minute600,000 a month
Analyst52Everything, including the model endpoints240 a minute1,000,000 a month