specification: API Commons Rate Limits
specificationVersion: '0.1'
schema: https://raw.githubusercontent.com/api-evangelist/interface-research/main/schema/api-commons.yml#/$defs/RateLimits
provider: APILayer
providerId: apilayer
generated: '2026-09-12'
method: searched
source: https://exchangerate.host/pricing, https://exchangeratesapi.io/pricing/,
  https://ipapi.com/pricing, openapi/_original/*.json (published 429 examples), and live
  header inspection of https://api.exchangerate.host/live and
  https://api.ipapi.com/api/check on 2026-09-12
created: '2026-05-04'
modified: '2026-09-12'
supersedes: >-
  A 2026-05-04 bulk-sweep scaffold that asserted 10/100/1000 requests per minute with
  X-RateLimit-Limit/Remaining/Reset and RateLimit-Policy response headers against
  api.apilayer.com. None of that was published by APILayer and none of those headers are sent.
headers:
  limit: null
  remaining: null
  reset: null
  retryAfter: null
  policy: null
  observed: []
  note: >-
    NO rate-limit response headers are sent. Live responses from api.exchangerate.host and
    api.ipapi.com on 2026-09-12 carried only date, content-type, cache-control, CORS, HSTS and
    Cloudflare headers - no X-RateLimit-*, no RateLimit-*, no Retry-After. This is the single
    most consequential gap for an agent: remaining quota is invisible at runtime, so the only
    way to discover exhaustion is to hit it.
responseCodes:
  throttled: 429
  quotaExceeded: 429
  serviceUnavailable: 503
  caveat: >-
    The 429 is documented in the OpenAPI. It was not verifiable live without burning a real
    quota, and the same surface was observed returning HTTP 200 for an authentication failure,
    so status-code fidelity on the error path is not established. Detect exhaustion from the
    body: success:false with error.type usage_limit_reached, daily_usage_limit_reached,
    fair_use_limit_reached or rate_limit_reached.
limit_kinds:
  - kind: monthly_quota
    published: true
    signal: error.type usage_limit_reached (code 104)
    note: The headline number on every pricing page. Resets on the billing month.
  - kind: daily_quota
    published: partial
    signal: error.type daily_usage_limit_reached (code 104)
    note: A daily cap exists - APILayer publishes the error for it - but no pricing page states
      the daily number for any tier.
  - kind: burst_rate
    published: false
    signal: error.type rate_limit_reached (code 106)
    note: >-
      APILayer's own error text says "You have exceeded the maximum rate limitation allowed on
      your subscription plan. Please refer to the Rate Limits section of the documentation."
      No such section is reachable - docs.apilayer.com renders client-side and no rate-limits
      page exists under any product path probed. The per-second/per-minute ceiling is therefore
      referenced by the API but published nowhere.
  - kind: fair_use
    published: false
    signal: error.type fair_use_limit_reached (code 104)
    note: An undocumented ceiling above the plan quota; the error directs the caller to support.
limits:
  - tier: Free
    scope: per-key (per product subscription)
    metric: requests
    limit: 100
    timeFrame: month
    burst: null
    applies: [ExchangeRate.host API, Exchange Rates API, ipapi]
    source: product pricing pages
  - tier: Basic (ExchangeRate.host / Exchange Rates API)
    scope: per-key
    metric: requests
    limit: 10000
    timeFrame: month
    burst: null
    overage_price_per_request: '0.005996'
    applies: [ExchangeRate.host API, Exchange Rates API]
  - tier: Professional / Professional Plus
    scope: per-key
    metric: requests
    limit: 100000
    timeFrame: month
    burst: null
    overage_price_per_request: '0.0023996'
    applies: [ExchangeRate.host API, Exchange Rates API]
  - tier: Business
    scope: per-key
    metric: requests
    limit: 500000
    timeFrame: month
    burst: null
    overage_price_per_request: '0.00079992'
    applies: [ExchangeRate.host API, Exchange Rates API]
  - tier: Standard (ipapi)
    scope: per-key
    metric: lookups
    limit: 50000
    timeFrame: month
    burst: null
    overage_price_per_request: '0.0010392'
    applies: [ipapi]
  - tier: Business (ipapi)
    scope: per-key
    metric: lookups
    limit: 500000
    timeFrame: month
    burst: null
    overage_price_per_request: '0.00047992'
    applies: [ipapi]
  - tier: Business Pro (ipapi)
    scope: per-key
    metric: lookups
    limit: 2000000
    timeFrame: month
    burst: null
    overage_price_per_request: '0.00019998'
    applies: [ipapi]
  - tier: Custom
    scope: contract
    metric: requests
    limit: -1
    timeFrame: month
    burst: null
    applies: [ExchangeRate.host API, Exchange Rates API, ipapi]
limit_count: 8
policies:
  - name: Overage, not cut-off
    description: Paid tiers meter past the quota at a published per-request price rather than
      hard-stopping, so exceeding the plan costs money instead of returning 429.
  - name: Backoff
    description: No Retry-After is sent, so a client must choose its own backoff. Distinguish
      rate_limit_reached (retry shortly) from usage_limit_reached (do not retry this month).
  - name: Scope
    description: Limits attach to the per-product access key, not to the APILayer account. Each
      subscribed product carries its own independent quota.
maintainers:
  - FN: Kin Lane
    email: info@apievangelist.com