REST API rate limits
All rate limits are per IP address.
Scope-specific limits override the default where listed. Some endpoints have tighter limits that override the defaults above. These are documented on the respective endpoint pages.
When the rate limit is exceeded, the API returns HTTP status
429 (Too Many Requests). Use exponential backoff before retrying (see Best practices below).
REST API error format
All v4 endpoints return errors as JSON. The format differs between public and private APIs.Public endpoints
Private endpoints
Authentication error reference
The following errors are returned by private REST endpoints when the request signature or credentials are invalid.HTTP status codes
Best practices
Exponential backoff
When rate limited, wait before retrying. Double the wait time after each failed attempt (1s → 2s → 4s → 8s).Batch requests
Combine multiple operations when the API supports batch endpoints. For example, use Bulk Limit Order instead of multiple single order requests.Nonce management
For private endpoints, ensure each request uses a unique, incrementing nonce. Use Unix timestamp in milliseconds whennonceWindow is enabled. Avoid concurrent requests with the same nonce.
Related resources
- API Reference Overview — Base URL and endpoint groups
- Authentication — Signing and header requirements
- WebSocket Rate Limits — WebSocket connection limits, error codes, and timeout behavior