Extend · Rate Limits

Extend Ai Rate Limits

The Extend API enforces account- and workspace-scoped throughput controls. Synchronous processor endpoints (/parse, /extract, /classify, /split) carry a five minute per-request timeout and are intended for onboarding and testing, while production traffic is expected to use the asynchronous run endpoints, batch endpoints, and webhooks. Workflow runs accept a priority value (default 50) that influences scheduling under load. Specific numeric request- and credit-per-minute limits are not publicly enumerated and are managed per account and plan, so this artifact is not reconciled.

Extend Ai Rate Limits is the machine-readable rate-limit profile for Extend on the APIs.io network, conforming to the API Commons Rate Limits specification.

It captures 5 rate-limit definitions, measuring seconds, requests, runs, priority, and bytes.

The profile also includes response codes documented for throttled.

Tagged areas include Document Processing, Document AI, Intelligent Document Processing, OCR, and Rate Limiting.

5 Limits Throttle: 429
Document ProcessingDocument AIIntelligent Document ProcessingOCRRate LimitingQuotasThrottling

Limits

Synchronous Request Timeout request
seconds
300
Sync /parse, /extract, /classify, /split fail after five minutes; use async runs.
Requests Per Minute account
requests
see provider documentation
Account- and workspace-scoped; not publicly enumerated.
Concurrent Runs account
runs
see provider documentation
Concurrency governed per account and plan; use batch endpoints for high volume.
Workflow Run Priority account
priority
50
Default priority value on workflow runs; higher priority is scheduled sooner under load.
Run Metadata Size request
bytes
10240
Custom run metadata is capped at roughly 10KB per run.

Sources