title: "API rate limits" description: "Minyu enforces explicit rate limits on API usage to ensure platform stability and fair usage across tenants." Rate limits apply to all API consumers and are e...
API rate limits
Minyu enforces explicit rate limits on API usage to ensure platform stability and fair usage across tenants. Rate limits apply to all API consumers and are enforced independently of usage-based credits.
Limits are evaluated continuously using rolling time windows and apply uniformly to all request types.
What is rate limited
Rate limits are evaluated per tenant and apply to:
- All API queries
- All API mutations
- All read and write operations
Rate limiting is independent of request shape, payload size, or operation type.
Limit units
The following limit units are enforced by default:
-
Requests per second
Limits short-term request bursts.
Default: 10 requests per second per tenant. -
Requests per hour
Limits sustained request volume over time.
Default: 10,000 requests per hour per tenant. -
Maximum execution time per request
Limits how long a single request may run.
Default: 5 seconds per request.
Execution time limits protect the platform from long-running or unbounded operations.
Error responses
When a rate limit is exceeded:
- The request is rejected
- No partial mutations are applied
- A rate-limit error response is returned
Short bursts above the limit may be tolerated, but sustained violations result in rejection. Repeated or prolonged violations may lead to temporary blocking.
Client behavior
Clients are expected to:
- Avoid tight polling loops
- Avoid repeated full-table reads
- Prefer filtered or incremental queries
- Batch writes sensibly
- Use notifications instead of polling where possible
Staying within rate limits does not guarantee sufficient usage credits, and available credits do not bypass rate limits. Both mechanisms operate independently.