Managing Model Endpoints using API You can use API to create, view, list, edit, describe, and delete model endpoints. Related informationCurl documentationPreparing to interact with the Cloudera AI Inference service APITo interact with Cloudera AI Inference service API, you need to obtain the domain name of the Cloudera AI Inference service and your Cloudera Data Platform JSON Web Token (JWT) and save it as environment variables.Creating a Model Endpoint using APIYou can select a specific Cloudera AI Inference service instance and a model version from AI Registry to create a new model endpoint.Listing Model Endpoints using APIConsider the following details for listing Model Endpoints using API.Describing a Model Endpoint using APIConsider the instructions for describing a Model Endpoint using API.Deleting a Model Endpoint using APIConsider the following instruction to delete a Model Endpoint using API.Autoscaling Model Endpoints using APIYou can configure the Model Endpoints deployed on Cloudera AI Inference service to auto-scale to zero instances when there is no load.Tuning auto-scaling sensitivity using the APITo customize the autos-caling sensitivity and requisites, set the target and metric fields in the autoscaling.autoscalingconfig parameter.Running Models on GPUFollow the guidelines for running Models on GPU.Deploying models with Canary deployment using APICloudera AI Inference service allows users to control traffic percentage to specific model deployments.
