Rate Limits
Rate Limits
| Endpoint Group | Limit | Window |
|---|---|---|
| /api/auth (login) | Class-based | five failures per account; separate per-IP credential and support-request abuse guards |
| /api/agents (autonomous) | Lower autonomous tier | per minute per actor |
| /api/v1/transactions | Up to ~120 req | per minute per API key |
| /api/v1/parse and /api/v1/validate | 120 req | per minute per API key |
| /api/v1/documents/outbound and /api/v1/documents/inbound | 120 req | per minute per API key |
| /api/v1/documents/batch | 60 req | per minute per API key (up to 100 documents per call) |
| /api/v1/documents/{documentId} and …/payload | 120 req | per minute per API key |
| /api/v1/document-types (public registry) | 300 req | per minute per client IP |
| /api/chat (AI assistant) | Up to ~30 req | per minute |
| /api/upload | 30 req per user, 90 req per network | per minute (in addition to the general /api/ budget) |
| All other /api/ routes | 300 req | per minute per actor |
Numbers above are typical ceilings; the actual cap that applies to your request is the lower of the per-endpoint ceiling and your plan-tier ceiling (see Acceptable Use). Rate-limit response headers (when present): X-RateLimit-Remaining and X-RateLimit-Reset (Unix timestamp). On a 429, retry after the time indicated in the response body or Retry-After header.
Still have questions?
Our support team typically responds within a few hours during business hours.
Contact support