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: AccelByte
providerId: accelbyte
created: '2026-07-01'
modified: '2026-07-01'
reconciled: false
tags:
- Gaming
- Backend
- BaaS
- Live Services
- Rate Limiting
- Quotas
- Throttling
description: >-
  AGS enforces per-environment, per-service rate limiting at the API gateway.
  Limits are applied per client/IP and per endpoint and are tuned per customer
  environment and plan; high-volume ingestion services (Game Telemetry) and
  authentication (IAM token) are the most commonly throttled surfaces. Exact
  numeric limits are environment-specific and not published as a single public
  table, so they are not reconciled in this artifact.
notes: >-
  Confirm per-service and per-environment limits with AccelByte during
  reconciliation; limits differ between the shared demo environment and
  provisioned customer environments and can be raised for enterprise plans.
sources:
- https://docs.accelbyte.io/
- https://accelbyte.io/products
responseCodes:
  throttled: 429
limits:
- name: IAM Token Requests
  scope: client
  metric: requests
  limit: see provider documentation
  notes: OAuth2 token issuance is rate limited per client to protect the auth service.
- name: Per-Service Requests
  scope: client
  metric: requests
  limit: see provider documentation
  notes: Each AGS service (Cloud Save, Statistics, Platform, etc.) has its own gateway limit.
- name: Game Telemetry Ingestion
  scope: client
  metric: events
  limit: see provider documentation
  notes: Telemetry endpoints accept batched events; ingestion is throttled per environment.
- name: Matchmaking Tickets
  scope: client
  metric: requests
  limit: see provider documentation
  notes: Ticket submission is limited to protect the matchmaking pipeline.
policies:
- name: Environment-Scoped Limits
  description: Limits are provisioned per customer environment and per namespace and can be increased on enterprise plans.
- name: Backoff Strategy
  description: Clients should implement exponential backoff with jitter and honor Retry-After on 429 responses.
- name: Batching
  description: Prefer batched/bulk endpoints (e.g., bulk stat updates, batched telemetry) to stay within request limits.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com