Model serving enables RHODS users to deploy trained models for testing and incorporation into intelligent applications. This functionality supports use cases for moving models into production applications. Deploying a model means making it available as a service accessible via an API to return predictions based on data inputs.

Reqs doc: https://docs.google.com/document/d/1TXLEyzpYX6inMHOlaUW8VbMxMvQ5Y9TEzHM87G9_1Yk/edit