OverviewQuickstartAPI referencePricingChangelogStatusGet an API key
DevelopersQuickstartAPI referenceConsolePricingAPI changelogService status
ExploreMoviesTV ShowsHelpTerms of ServicePrivacy PolicyGet the appCookie PolicyPricingRefunds and CancellationAcceptable Use PolicyRights holdersService statusCheck a takedown noticeWhat's new
·

© 2026 Wambesha

v0.110.0
Overview

Discovery

  • SearchGET
  • DiscoverGET
  • LookupGET
  • GenresGET
  • Similar titlesGET

Titles

  • Batch titlesGET
  • Title recordGET
  • SeasonsGET
  • EpisodesGET
  • CreditsGET
  • ImagesGET
  • Where a title is heldGET

People

  • PersonGET
  • FilmographyGET

Changes

  • Change feedGET
  • ChangelogGET

Account

  • Plan and usageGET
  • Usage historyGET
  • Recent requestsGET
  • KeysGET
  • Rotate keyPOST

21 of 21 endpoints

Account/usage

GET/v1/account/usage

Calls quota only

Requests, verifications, errors and latency over the last day, week or month, broken down by endpoint and key. Never charged.

Parameters

  • range24h | 7d | 30doptionaldefault 24h

    24h (default), 7d or 30d.

    More⌄

    An unknown value is a 400 naming the accepted set, never a silent fallback to the default. Case-insensitive; shown here in lowercase.

Response fields

  • range24h | 7d | 30d
  • fromstring · date-time
  • tostring · date-time
  • bucketsUsageBucket[]
  • byEndpointEndpointUsage[]
  • errorGroupsErrorGroup[]
  • perKeyKeyUsage[]
  • baselineobject · UsageBaseline

    The immediately preceding span of EQUAL length, so a number has something to be compared against.

Response headers

  • X-Request-ID

    The id of this request, on every response - success or error. Quote it to support: it is how we find your request in our logs. An X-Request-ID you send is honoured and echoed back, so you can use your own; otherwise we generate one. /v1 error bodies carry no trace id of their own.

  • X-Quota-Reset

    RFC 3339 instant at which the billing period rolls. Present on every response to a request whose key was metered - every 2xx, every error a handler returned, and the quota 429. Absent on a refusal that happens before metering: a missing or unknown key (401), a restricted key or a plan that does not carry the endpoint (403), a public demo-key gate refusal, and the pre-authentication flood throttle.

  • X-Calls-Remaining

    Requests left in the period after this one, never below 0. Present whenever X-Quota-Reset is (see there for when a response carries no quota headers).

  • X-Verifications-Remaining

    How many verifications you can still make before a limit stops you. The daily limit and the monthly allowance are counted separately; this is whichever of the two runs out first. Sent whenever X-Quota-Reset is.

  • X-Quota-Grace

    true when THIS request fell inside the 10% grace band above a monthly ceiling. A grace request is served and is not billed; treat true as "back off now", not as headroom. Always true or false, never absent when the other quota headers are present.

  • Retry-After429

    Seconds to wait before retrying. On every 429 - the quota hard stop, the public demo key's limits, and the pre-authentication flood throttle (which carries this header alone, because no key has been read yet).

  • X-Demo-Challenge-ModeDemo key only

    Sent only on responses to the public demo key. The value interactive asks the documentation page to show the visible verification widget - it is sent once a browser has made more than five runs in ten minutes or has received a 429. The page then repeats the header, X-Demo-Challenge-Mode: interactive, on its next request, with the widget's token in X-Demo-Challenge. Requests made with your own key never receive it.

Errors

  • validation-failed

  • unauthenticated

  • forbidden

  • plan-required

  • validation-failed

  • rate-limited

  • upstream-unavailable

Details⌄

Per-account and per-key buckets: requests by status class, errors grouped by cause with an example and a documentation link, latency percentiles, verifications spent, and a per-endpoint breakdown. Each range carries the immediately preceding equal-length span as a baseline.

Percentiles come from a stored bucket ladder, so each is an UPPER BOUND rather than an interpolation within a bucket - the honest reading of bucketed data.

No cache/origin split and no cache-share metric. Our caching is our cost to manage, not a signal to optimise against.

PreviousPlan and usageNextRecent requests

Try it

Loading the demo key…

24h (default), 7d or 30d.

curl "https://api.wambesha.com/v1/account/usage" \
  -H "Authorization: Bearer YOUR_API_KEY"

Example

{
  "range": "24h",
  "from": "2026-09-26T16:21:33Z",
  "to": "2026-09-27T16:21:33Z",
  "buckets": [],
  "byEndpoint": [],
  "errorGroups": [],
  "perKey": [],
  "baseline": {
    "from": "2026-09-25T16:21:33Z",
    "to": "2026-09-26T16:21:33Z",
    "requests": 0,
    "verifications": 0,
    "errorRatePercent": 0
  }
}