generated: '2026-07-19'
method: searched
source: https://docs.gc.ai/api-reference/concepts/rate-limits
docs: https://docs.gc.ai/api-reference/concepts/rate-limits
scope: per-organization
notes: >-
  Limits apply per organization and per API key; the organization limit is the
  binding constraint. Throttled requests return 429 with a machine code
  RATE_LIMITED. Honor Retry-After; use the async job pattern with paced
  enqueueing for batch workloads.
throttle:
  status: 429
  error_code: RATE_LIMITED
headers:
  - name: Retry-After
    description: Seconds to wait before retrying.
  - name: RateLimit-Limit
    description: Quota for the current window.
  - name: RateLimit-Remaining
    description: Requests left in the current window.
  - name: RateLimit-Reset
    description: Seconds until the quota resets.
limits:
  - tier: inference
    applies_to: [createChatCompletion, runPlaybook]
    limit_count: 1
    window: 1 minute
    burst: 3
    window_seconds: 180
    description: >-
      Chat completions and playbook runs: ~1 request/minute per organization
      with a burst of 3 over a rolling 180-second window.
  - tier: standard
    applies_to: CRUD operations (files, folders, projects, playbooks, skills, profiles)
    limit_count: 120
    window: 1 minute
    description: "Everything else: 120 requests/minute per organization."