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: Axiom
providerId: axiom-co
created: '2026-07-02'
modified: '2026-07-02'
reconciled: false
tags:
- Observability
- Log Management
- Event Data
- Rate Limiting
- Quotas
description: >-
  Axiom applies three distinct categories of API limit and communicates each
  through its own set of response headers. General API rate limits cap requests
  per minute at either the user or organization scope. Query limits are metered
  in gigabyte-milliseconds (GB*ms) of query compute against your plan. Ingest
  limits are metered in bytes per month against your plan, and Axiom
  additionally monitors ingest request rate - an organization sending an
  excessive, sustained number of requests per second far above normal usage may
  have its ingest suspended to protect system stability. When any limit is
  exceeded the API returns HTTP 429 Too Many Requests with a JSON body of
  {"message": "rate limit exceeded"}. Axiom recommends batching clients to
  aggregate events and reduce request rate.
notes: >-
  Specific numeric per-minute request ceilings vary by plan and scope and are
  surfaced at runtime in the X-RateLimit-* response headers rather than as a
  single published constant. Query and ingest ceilings track your plan's monthly
  allowances. Verify current values on the Axiom API limits page during
  reconciliation.
sources:
- https://axiom.co/docs/restapi/api-limits
- https://axiom.co/docs/restapi/introduction
- https://axiom.co/docs/restapi/ingest
responseCodes:
  throttled: 429
  throttledBody: '{"message": "rate limit exceeded"}'
headers:
  general:
  - name: X-RateLimit-Scope
    description: Whether the limit applies at the user or organization scope.
  - name: X-RateLimit-Limit
    description: Maximum number of requests allowed per minute in the current window.
  - name: X-RateLimit-Remaining
    description: Requests remaining in the current window.
  - name: X-RateLimit-Reset
    description: UTC epoch seconds at which the current window resets.
  query:
  - name: X-QueryLimit-Limit
    description: Plan query-compute ceiling expressed in gigabyte-milliseconds (GB*ms).
  - name: X-QueryLimit-Remaining
    description: Remaining query compute in GB*ms.
  - name: X-QueryLimit-Reset
    description: UTC epoch seconds at which the query allowance resets.
  ingest:
  - name: X-IngestLimit-Limit
    description: Maximum bytes that can be ingested per month under the plan.
  - name: X-IngestLimit-Remaining
    description: Bytes of ingest allowance remaining.
  - name: X-IngestLimit-Reset
    description: UTC epoch seconds at which the ingest allowance resets.
limits:
- name: General API Requests
  scope: user-or-organization
  metric: requests
  limit: per-minute (surfaced in X-RateLimit-Limit)
  notes: Applies to standard REST API operations; scope reported in X-RateLimit-Scope.
- name: Query Compute
  scope: organization
  metric: gb_milliseconds
  limit: per plan (surfaced in X-QueryLimit-Limit)
  notes: APL and legacy queries consume query compute measured in GB*ms against the plan.
- name: Ingest Volume
  scope: organization
  metric: bytes
  limit: per month per plan (surfaced in X-IngestLimit-Limit)
  notes: Monthly data-loaded allowance; overage and rates depend on plan tier.
- name: Ingest Request Rate
  scope: organization
  metric: requests_per_second
  limit: monitored (no fixed published constant)
  notes: >-
    Sustained excessive requests-per-second far above normal usage may lead to
    ingest suspension. Use batching clients to aggregate events.
policies:
- name: Batching
  description: >-
    Axiom recommends batching clients that aggregate events before sending to
    stay within ingest request-rate expectations and reduce 429s.
- name: Backoff Strategy
  description: >-
    Clients should implement exponential backoff with jitter on 429 responses
    and read the X-RateLimit-Reset / X-QueryLimit-Reset / X-IngestLimit-Reset
    headers to time retries.
- name: Header-Driven Throttling
  description: >-
    Remaining allowance for each category (general, query, ingest) is exposed in
    dedicated response headers so clients can self-throttle before hitting 429.
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com