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: ENTSO-E
providerId: entso-e
created: '2026-07-11'
modified: '2026-07-11'
reconciled: true
tags:
- Electricity
- Energy Markets
- Day-Ahead Prices
- Balancing
- Rate Limiting
- Quotas
description: >-
  The ENTSO-E Transparency Platform publishes explicit, numeric limits in its
  RESTful API user guide. Request rate is capped at 400 requests per minute,
  counted per IP address AND per security token; crossing the cap earns a
  temporary 10-minute ban, during which requests receive HTTP 429 with the
  message "Max allowed requests per minute from each unique IP is max up to
  400 only". Independent of rate, each query is bounded in size - standard
  (table) data items such as day-ahead prices, load, and generation allow at
  most a one-year range per request, while document-based (curve) items such
  as outages and capacity allocations return at most 100 documents per
  request (200 for outages), with an offset parameter (0-4800) on some items
  allowing paging up to 4,900 documents. Over-limit queries are rejected with
  an Acknowledgement_MarketDocument explaining the reason.
notes: >-
  When a query is rejected for exceeding limits, the guide recommends
  splitting the time interval into smaller consecutive windows or adding
  optional selection criteria (psrType, businessType). Empty results are also
  delivered as an Acknowledgement_MarketDocument with Reason code 999 "No
  matching data found" - treat it as no-data, not as an error to retry.
sources:
- https://transparency.entsoe.eu/content/static_content/Static%20content/web%20api/Guide.html
- https://transparencyplatform.zendesk.com/hc/en-us
responseCodes:
  throttled: 429
limits:
- name: API Request Rate
  scope: ip-address-and-token
  metric: requests
  limit: 400 per minute
  notes: >-
    Counted per unique IP address and per security token. Exceeding the cap
    results in a temporary ban of 10 minutes; banned requests receive HTTP
    429 Too Many Requests.
- name: Standard Data Item Query Range
  scope: request
  metric: time-range
  limit: one year
  notes: >-
    Applies to table-style data items (day-ahead prices A44, load A65,
    generation A75, balancing A82-A86, flows A11, and most others). Longer
    ranges are rejected with an Acknowledgement_MarketDocument.
- name: Document-Based Query Size
  scope: request
  metric: documents
  limit: 100
  notes: >-
    Curve-type data items (capacity allocations, contracted reserves,
    procured balancing capacity) return at most 100 published documents per
    request; one platform table row corresponds to one document.
- name: Outage Query Size
  scope: request
  metric: documents
  limit: 200
  notes: The document limit is raised to 200 per request for outage (unavailability) data items.
- name: Offset Paging
  scope: request
  metric: documents
  limit: offset 0-4800 (max 4,900 documents per query)
  notes: >-
    Items that support the offset parameter page in steps of the document
    limit; offset=n returns documents n+1 onward, restricted so a query can
    reach at most 4,900 documents.
policies:
- name: Temporary Ban
  description: >-
    Exceeding 400 requests/minute triggers a 10-minute ban keyed to the IP
    address / security token, served as HTTP 429.
- name: Rejection as Acknowledgement Document
  description: >-
    Requests over the range or document limits, with missing/duplicate
    parameters, or matching no data are answered with an IEC 62325
    Acknowledgement_MarketDocument carrying a Reason code (999 = no matching
    data found) rather than a JSON error.
- name: Backoff Strategy
  description: >-
    Clients should self-throttle below 400 requests/minute, wait out the full
    10-minute ban on 429 (retrying during the ban extends nothing), and split
    large historical pulls into one-year (or smaller) consecutive windows.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com