This document describe the limits and quota of requesting API.
We employ state of the art containerized orchestration system to host our services on top of cloud computing power that can adjust scale according to the load. User's requests are distributed by redundant load balancer among multiple availability zones, then distributed further to one of the service host instances. Every day dozens of developers make requests to the Nodeflux. To help manage the sheer volume of these requests, limits are placed on the number of requests that can be made. These limits help us provide the reliable and scalable API that our developer community relies on and to ensure the equitability of the system resource.
The following table lists the default Query Per Second (QPS) for the IdentifAI service per second.
- Request with multiple images (e.g. face match with liveness), and bigger image file size might have longer latency.
- The latency above is only computation latency (excluding data transfer).
- The QPS limits are projected up to the specified caps. It doesn't guarantee the availability of the quotas displayed above.