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: Mindee
providerId: mindee
created: '2026-05-25'
modified: '2026-05-25'
reconciled: true
tags:
  - Document Parsing
  - OCR
  - Rate Limiting
  - Quotas
description: Reconciled limits for the Mindee V2 inference API. Mindee meters consumption by credits rather than per-second request rate; technical limits cover file size, page count, and asynchronous job behavior.
sources:
  - https://docs.mindee.com/integrations/technical-limitations
  - https://docs.mindee.com/integrations/response-times
  - https://docs.mindee.com/integrations/polling-for-results
responseCodes:
  throttled: 429
  quotaExceeded: 402
  badRequest: 400
  unauthorized: 401
  notFound: 404
  payloadTooLarge: 413
algorithm: credit-bucket
quotas:
  - dimension: monthly credits
    description: Each plan ships with a fixed monthly credit allowance; overage is billed per-credit at the plan's overage rate.
  - dimension: file size
    description: Files up to 100 MB per inference request.
  - dimension: page count
    description: PDFs up to 200 pages; larger documents must be split first.
  - dimension: concurrency
    description: Inference is asynchronous; clients enqueue a job and poll for results or receive a webhook callback.
notes:
  - Mindee does not publish a per-second request-rate limit; throughput is capped by the asynchronous job pipeline and the plan's monthly credit budget.
  - Enterprise customers can negotiate higher concurrency and dedicated processing capacity under a custom SLA.
  - Asynchronous jobs typically complete within a few seconds for single-page documents; larger documents and complex models take longer.