Private AI is primarily designed to be self-hosted by the user via a container, to provide users with the best possible experience in terms of latency and security. It would be counter-productive to send sensitive data across the Internet to a 3rd party system for the purpose of preserving privacy. It also ensures that Private AI never sees or handles customer data, unlike cloud APIs which usually retain a right to use any data passed through the system for service improvements and ML model development.
Instead of running the container locally, it is also possible to use a cloud version of the API, hosted at the following endpoint: https://api.private-ai.com/deid/v1
Custom integrations that do not rely on Docker can also be delivered upon request.
Installation is organized as follows:
- Prerequisites and System Requirements details the minimum and recommended hardware to run the container on.
- Grabbing the Image describes how to load the Private AI image.
- Running the Container covers how to create and run the container on your local machine.
- Deploying into Production describes how to run the container in production with orchestrators such as AWS ECS.
- Kubernetes Setup Guide is a self-contained guide on how to run the container using Kubernetes.
- Benchmarks provides some performance numbers on the CPU and GPU containers.
- Concurrency talks about recommended concurrency settings.