generated: '2026-08-13'
method: searched
source: https://docs.sumble.com/api/api
limit_count: 1
rate_limits:
  - scope: per-user
    applies_to: all endpoints (aggregated across the whole API surface)
    limit: 10
    unit: requests
    window: 1s
    burst: >-
      Occasional bursts are allowed; the documented text is "Each user can do
      10 requests per second (aggregated through all the endpoints), allowing
      occasional bursts." No burst size is published.
    exceeded_status: 429
    docs: https://docs.sumble.com/api/api#rate-limits
response_headers:
  documented: []
  observed:
    - name: api_version
      example: v9
      note: >-
        Observed on a live unauthenticated POST to
        https://api.sumble.com/v9/organizations (HTTP 401). Echoes the API
        version the request resolved to; it is not a rate-limit signal.
  rate_limit_headers: none
  retry_after: not documented
  note: >-
    Sumble publishes NO rate-limit response headers — no X-RateLimit-*, no
    RateLimit-* (RFC 9331 draft), no Retry-After. An agent cannot read its
    remaining budget from a response; it can only detect exhaustion after the
    fact from the 429 status. This is the runtime signal gap for this
    provider.
quotas:
  model: credit-based
  note: >-
    Separate from rate limiting, every call consumes account credits and a
    402 is returned on insufficient credits. Per-endpoint credit costs are
    published at https://docs.sumble.com/api/api#credits and mirrored in
    plans/sumble-plans-pricing.yml.
  exhausted_status: 402
evidence:
  - {url: 'https://docs.sumble.com/api/api.md', status: 200, kind: docs}
  - {url: 'https://api.sumble.com/v9/organizations', status: 401, kind: live-probe, note: header inspection}
checked: '2026-08-13'