Rate limits
How Plurall AI enforces fair use across keys and tenants.
Each environment carries its own API-key, organization-minute, and organization-day limits. Plurall AI returns `RateLimit-Limit`, `RateLimit-Remaining`, `RateLimit-Reset`, and a 429 with `Retry-After` when you exceed the active bucket.
Tiers
Sandbox defaults to 120 requests per API key per minute and 5,000 requests per organization per day. Staging and production have higher defaults and can be tuned through environment-level policy without changing client code.
Headers
`RateLimit-Limit` is the active bucket size, `RateLimit-Remaining` is the most constrained remaining bucket, `RateLimit-Reset` is the reset epoch in seconds, and `X-RateLimit-Policy` identifies whether the API key or organization bucket is currently most constrained.