specification: API Commons Rate Limits
specificationVersion: '0.1'
provider: Reducto
providerId: reducto
created: '2026-06-12'
modified: '2026-06-12'
url: https://reducto.ai/pricing
retryAfter: true
throttled:
  - statusCode: 429
    description: >
      Returned when the rate limit for API calls per second is exceeded.
      The Reducto Python and Node.js SDKs automatically retry on 429 responses
      using exponential backoff with up to 2 retries by default.
limits:
  - scope: Standard Plan
    metric: requests_per_second
    limit: 1
    timeFrame: second
    description: >
      Standard (free/pay-as-you-go) plan is limited to 1 API call per second
      across all endpoints.
  - scope: Growth Plan
    metric: requests_per_second
    limit: 10
    timeFrame: second
    description: >
      Growth plan supports up to 10 API calls per second with up to 5
      concurrent active priority requests.
  - scope: Growth Plan
    metric: concurrent_priority_requests
    limit: 5
    timeFrame: concurrent
    description: >
      Maximum concurrent priority requests for the Growth plan.
  - scope: Enterprise Plan
    metric: requests_per_second
    limit: 100
    timeFrame: second
    description: >
      Enterprise plan supports 100+ API calls per second with custom
      concurrency limits negotiated per deployment.
notes:
  - >
    Async operations (async-parse, async-extract, etc.) consume rate limit
    quota at submission time and can be polled via job ID without additional
    quota consumption.
  - >
    The Reducto Python SDK and Node.js SDK default to 2 automatic retries
    with exponential backoff for 429 and 5xx errors.
  - >
    Default request timeout is 1 hour to accommodate large document processing
    jobs.