Rate limits and usage

Per-key rate limits, the headers that report them, and the limits of each plan.

Rate limits

Each key is limited to its own number of calls a minute, shared by every server that holds the key. The limit is set when the key is made, from 1 to 1,000, and is 60 by default. A call over the limit is answered 429 with the code rate_limited and a Retry-After header in seconds.

Rate limit headers
HeaderMeaning
X-RateLimit-LimitCalls a minute the key is allowed.
X-RateLimit-RemainingCalls left in the current minute.
X-RateLimit-ResetSeconds until the minute resets.
Retry-AfterOn a 429 only: seconds to wait before the next call.

A call with a key that is wrong, revoked or expired is counted against the network address that sent it, and an address that sends many is slowed down.

Limits of each plan

The rate limit belongs to the key and is the same on every plan that includes the API: 1 to 1,000 calls a minute, 60 by default. What differs by plan is the allowance a call uses. These are the figures the API enforces.

Limits of each plan
PlanAPITasksLargest fileFiles in one runAI credits a monthSignature requests a month
FreeNot included3 a day25 MB5None2
ProNot included2,000 a month100 MB2550050
TeamIncluded10,000 a month250 MB502,500300
BusinessIncluded50,000 a month500 MB10010,0001,500
EnterpriseIncluded100,000 a month1000 MB20020,0005,000

Enterprise limits are agreed with each customer, and the figures above are where they start. A limit can also be changed for one account by PDF Loft.

Usage and plan limits

API calls are counted in the same shared task pool as the web apps and the mobile and desktop apps. One task is one output file produced by a tool, as everywhere in PDF Loft. A run reserves its tasks when it starts, commits them when it succeeds, and gives them back when it fails or is cancelled. When the pool is used, a run is refused with tasks.exhausted and the details name the reset time. The largest file and the files per run follow the plan of the account.