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: Riot Games
providerId: riot-games
created: '2026-05-04'
# Provenance stamped 2026-08-11: this artifact was written by the API Evangelist
# bulk sweep dated 2026-05-04, not harvested from the provider. See roadmap#35.
method: generated
modified: '2026-05-05'
reconciled: true
tags:
  - Riot Games
  - League of Legends
  - VALORANT
  - Rate Limiting
description: >-
  Riot enforces app-level rate limits per region and per API key, plus method-level limits on
  individual endpoints. Limits are applied across both the user's personal/production app
  and the underlying service; either layer can return 429. Limits scale with key tier
  (development/personal vs production), with production-key limits expandable for developers
  in good standing.
sources:
  - https://developer.riotgames.com/docs/portal
  - https://developer.riotgames.com/api-keys.html
headers:
  retryAfter: Retry-After
  rateLimitType: X-Rate-Limit-Type
responseCodes:
  throttled: 429
limits:
  - name: Development / Personal key burst
    scope: key/region
    metric: requests_per_second
    limit: 20
    timeFrame: second
    notes: >-
      Personal and development keys share this 20 req / 1 s burst limit, enforced per region.
  - name: Development / Personal key sustained
    scope: key/region
    metric: requests_per_2_minutes
    limit: 100
    notes: >-
      100 requests per 2 minutes, per region, on personal/development keys.
  - name: Production key burst
    scope: key/region
    metric: requests_per_10_seconds
    limit: 500
    notes: >-
      500 requests per 10 seconds, per region. Expandable for approved production apps in
      good standing.
  - name: Production key sustained
    scope: key/region
    metric: requests_per_10_minutes
    limit: 30000
    notes: >-
      30,000 requests per 10 minutes, per region. Expandable on request.
  - name: Method-level limits
    scope: key/method/region
    metric: varies
    limit: see endpoint reference
    notes: >-
      Individual endpoints carry their own per-method limits in addition to the app-level
      caps; method limits surface as 429 with X-Rate-Limit-Type when triggered at the edge.
policies:
  - name: Per-Region Enforcement
    description: >-
      All rate limits are enforced per regional shard (e.g. NA1, EUW1, KR). A single key can
      run against multiple regions but each region carries its own counter.
  - name: Retry-After Compliance
    description: >-
      Clients must honor the Retry-After header on 429 responses and pause until the
      indicated time before retrying.
  - name: Rate Limit Type Header
    description: >-
      When the 429 originates at the API edge, X-Rate-Limit-Type identifies whether the
      breach was at the application or method layer. The header is not guaranteed for 429s
      surfaced by the underlying service.
  - name: Production Approval
    description: >-
      Production-key issuance and any subsequent rate-limit raise requires Developer
      Relations review of a working prototype and adherence to Riot's developer policies.
maintainers:
  - FN: Kin Lane
    email: kin@apievangelist.com