Models API

Train, deploy, and run inference on custom models.