generated: '2026-08-13'
method: searched
source: https://docs.talkable.com/api_v2/intro/
docs: https://docs.talkable.com/api_v2/intro/
api: Talkable REST API v2
limit_count: 0
note: >-
  Talkable documents that it throttles, and documents the runtime signal and the
  client obligation, but deliberately publishes NO numeric limits. From the API
  introduction: "Rate limits are not published because the computation logic is
  evolving continuously to maximize reliability and performance for customers."
  limit_count is therefore an honest 0 — the policy exists and is documented, the
  numbers do not. Talkable also states it does not publish rate-limit response
  headers; the only runtime signal an agent gets is the 429 status code.
policy:
  documented: true
  scope: per-site (customer)
  model: >-
    Talkable computes the resources each client consumes across ALL requests from
    all integrations — not only REST API v2 calls, but also the JavaScript
    integration library, the mobile SDKs, and any other integration. When a site's
    consumption rises unexpectedly, Talkable begins throttling that site.
  window: unpublished
  limit: unpublished
  burst: unpublished
limits: []
response:
  status_on_exhaustion: 429
  status_name: Too Many Requests
  headers: []
  headers_note: >-
    No X-RateLimit-*, RateLimit-* or Retry-After headers are documented for
    Talkable API v2. An agent must treat the bare 429 as the whole signal.
  retry_guidance: >-
    Talkable recommends retrying on 429. Consecutive calls may keep returning 429
    until server load drops. Clients should implement retry with incremental
    backoff "in the realm of seconds and up to minutes".
related:
  errors: errors/talkable-problem-types.yml
  conventions: conventions/talkable-conventions.yml
x-evidence:
  - url: https://docs.talkable.com/api_v2/intro.md
    http_status: 200
    fetched: '2026-08-13'
    section: Request Throttling