Based on the Permit.io documentation, here's what's available regarding latency, SLA, and performance: Latency The PDP (Policy Decision Point) provides "fast, zero-latency, scalable, secure policy decision making" (1) . When deployed as a local sidecar, you get "zero-latency between your application and the PDP" . The documentation indicates PDPs answer queries "in ≤ ms" (milliseconds or less) . SLA (Service Level Agreement) The SLA varies by pricing tier (2) : • ​*Free & Growth tiers*​: Best Effort support SLA • ​*Scale tier*​: Best Effort support SLA, with 99.9% (0.999) Cloud Uptime • ​*Enterprise tier*​: Enterprise SLA Document, with 99.95% (0.9995) Cloud Uptime (upgradeable to 99.99%) For Local PDP Uptime, all tiers offer 99.9999% (0.999999) availability . TPS (Transactions Per Second) The documentation doesn't specify exact TPS numbers. However, it mentions that PDPs support "bulk checks" for batching multiple authorization queries (3) , and the architecture is designed to be "scalable" with the ability to run PDPs as sidecars that scale horizontally with your application (1) (4) . For production deployments, Permit.io recommends running a local PDP inside your network to "minimize network latency and ensure high availability" (3) .

To learn more, see the sources I considered: [[1] at Scale](https://docs.permit.io/how-to/deploy/overview) [[2] Pricing Packages and SaaS Models | Permit.io](https://www.permit.io/pricing) [[3] Permit's PDP (Policy Decision Point)](https://docs.permit.io/concepts/pdp/overview) [[4] Best Practices for Microservice Authorization](https://www.permit.io/blog/best-practices-for-authorization-in-microservices)