Rate limits
The Orb API imposes safeguards to help ensure service availability against bursts of incoming traffic. Orb is implemented in the critical path for many of our customers and our rate limits are designed with that consideration in mind:
| Category | Applicable endpoints | Live mode limit | Test mode limit | Trial limit |
|---|---|---|---|---|
| Analytics | Subscription usage, subscription costs, and the customer costs endpoints. | 10 queries per second | 5 queries per second | 1 query per second |
| Ledger | Customer credit balance, and the customer credit balance by external ID endpoints. | 10 queries per second | 5 queries per second | 1 query per second |
| Read-only queries | All GET requests exclusive of the Analytics scope. | 100 queries per second | 50 queries per second | 10 queries per second |
| Write actions | All POST, PUT, PATCH, and DELETE requests, excluding the ingest event endpoint. | 50 operations per second | 5 operations per second | 1 operation per second |
| Ingestion requests | Ingest | 100 operations per second | 10 operations per second | 2 operations per second |
| Ingestion volume | Ingest | 10,000 events ingested per minute. This quota can be increased significantly upon request. For information regarding support for higher event ingestion volumes, see the guide on high throughput ingestion. | 2,000 events ingested per minute | 100 events ingested per minute |
If your workflow requires additional capacity, please reach out to the Orb team to help provision dedicated resources for your workload.
Handling rate limits
In the event of a 429 Too many requests error, we recommend you retry the request with an exponential backoff.