specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
generated: '2026-08-13'
method: searched
source: https://worldnewsapi.com/docs/quotas-and-rate-limiting/
provider: World News API
providerId: worldnewsapi
created: '2026-07-11'
modified: '2026-08-13'
reconciled: true
reconciledOn: '2026-08-13'
limit_count: 6
tags:
- News
- News Search
- Rate Limiting
- Quotas
- Points
description: >-
  World News API enforces two overlapping controls - a daily point (credit)
  allowance and per-request rate limits (requests per second and maximum
  concurrent requests). Every endpoint consumes a number of points that count
  against the plan's daily allowance; exceeding the allowance blocks calls on the
  Free plan or bills per additional point on paid plans. Rate and concurrency
  limits scale up with plan tier - Free is 1 request/sec and 1 concurrent
  request; Reporter 2/sec and 5 concurrent; Journalist 10/sec and 10 concurrent;
  Editor 20/sec and 20 concurrent.
notes: >-
  Re-read from the provider's own rate-limiting documentation and pricing page
  on 2026-08-13. Baseline cost is 1 point per request plus 0.01 points per
  result returned, with per-endpoint exceptions stated at the bottom of each
  endpoint's documentation page. Quota resets at midnight UTC. The provider
  publishes its budget signal on proprietary X-API-Quota-* response headers
  rather than the IETF RateLimit-* fields, and sends no Retry-After on 429.
sources:
- https://worldnewsapi.com/docs/quotas-and-rate-limiting/
- https://worldnewsapi.com/pricing/
- https://worldnewsapi.com/docs/
responseCodes:
  quotaExceeded: 402
  throttled: 429
  unauthorized: 401
  planFeatureNotIncluded: 403
responseHeaders:
- name: X-API-Quota-Request
  meaning: Points consumed by this request.
  returned_on: every successful response
- name: X-API-Quota-Used
  meaning: Points consumed today in total; resets to zero at midnight UTC.
  returned_on: every successful response
- name: X-API-Quota-Left
  meaning: Points remaining today under the current plan.
  returned_on: every successful response
- name: Retry-After
  meaning: Not returned. The provider documents no Retry-After on 429; use exponential backoff.
  returned_on: never
standardHeaders:
  ratelimit_draft: false
  x_ratelimit: false
  note: >-
    Neither the RateLimit-* (IETF draft) nor the X-RateLimit-* families are
    used. An agent must read the X-API-Quota-* headers to know its budget.
discrepancies:
- field: Free plan request rate
  pricing_page: 1 request per second
  rate_limit_docs: 60 requests in 1 minute
  note: >-
    The two provider pages disagree on the shape of the Free limit (a per-second
    cap versus a per-minute budget). Both readings are recorded rather than
    resolved; the per-minute form permits short bursts the per-second form does
    not.
- field: Editor plan concurrency
  pricing_page: 20 concurrent requests
  rate_limit_docs: 10 requests simultaneously
  note: The pricing page and the rate-limiting guide state different concurrency ceilings for the top tier.
limits:
- name: Daily Points (Free)
  scope: account
  metric: points
  limit: 50 per day
  notes: After 50 points, no further calls are allowed until the next day.
- name: Daily Points (Reporter)
  scope: account
  metric: points
  limit: 500 per day included, then $0.003 per additional point
  notes: Paid overage past the included daily allowance.
- name: Daily Points (Journalist)
  scope: account
  metric: points
  limit: 5000 per day included, then $0.0025 per additional point
  notes: Paid overage past the included daily allowance.
- name: Daily Points (Editor)
  scope: account
  metric: points
  limit: 50000 per day included, then $0.002 per additional point
  notes: Paid overage past the included daily allowance.
- name: Requests per Second
  scope: account
  metric: requests
  window: second
  limit: 1 (Free) / 2 (Reporter) / 10 (Journalist) / 20 (Editor)
  notes: >-
    Per-second request cap scales with plan tier. The rate-limiting guide states
    the Free limit as 60 requests in 1 minute instead; see discrepancies above.
- name: Concurrent Requests
  scope: account
  metric: connections
  limit: 1 (Free) / 5 (Reporter) / 10 (Journalist) / 20 (Editor)
  notes: >-
    Maximum number of in-flight requests at once, scales with plan tier. The
    rate-limiting guide states 10 rather than 20 for Editor.
policies:
- name: Point Metering
  description: Each endpoint call consumes a plan-independent number of points based on the endpoint and parameters; points are the primary quota.
- name: Backoff Strategy
  description: Honor 429 (rate limited) and 402 (quota exceeded) responses with exponential backoff and by staying within the plan's per-second and concurrency limits.
- name: Quota Headers
  description: Inspect X-API-Quota-Request, X-API-Quota-Used, and X-API-Quota-Left response headers to track remaining daily points.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com