minimusdocs
Help center

Rate limits

How many requests a key or an AI app may make, and what happens past that.

Your panel limits how fast each API key and each connected AI app may call it, so one runaway script never slows the panel down for everyone else, including the dashboard and your customers' players.

  • 120 calls a minute per key, and per AI app, unless the panel's owner changed it: Settings → Panel settings → Minimus API → Calls a minute per API key (10 to 6,000).
  • The count is the panel's, across all its processes: two copies of a script using the same key share one limit.
  • A call refused for going too fast does not count, so a client that waits gets back in.
  • Wrong keys count as failed sign-ins: an address that keeps sending wrong or revoked keys is blocked for a while by the panel's attack protection, like one guessing passwords (429 for every call from it, even with a good key).

Past the limit

The panel answers 429 Too Many Requests, does nothing, and says how long to wait:

HTTP/1.1 429 Too Many Requests
Retry-After: 12
Content-Type: application/json

{"detail": "Request was throttled. Expected available in 12 seconds."}

Wait the number of seconds in Retry-After, then send the same request again.

Staying under it

  • Ask for bigger pages instead of many small ones: page_size goes up to 200. See Requests and lists.
  • Change many things in one call where the API has one: mass edits of streams, bulk actions on lines and episodes run as one background task instead of hundreds of calls.
  • Don't poll fast. Checking a background task every few seconds is plenty.
  • One key per script, so two busy scripts don't share one limit.

The playlist editor's API

The playlist editor's own token API has its own limit: 120 requests a minute per token, and a few minutes' pause for an address after 10 wrong tokens. See Playlist editor API.