An ML engineer must use AWS CloudFormation to create the ML model that an Amazon SageMaker endpoint will host. Which CloudFormation resource should the engineer declare to represent that model?

The AWS::SageMaker::Model CloudFormation resource is the correct choice because it represents the machine learning model itself, including its artifacts (e.g., S3 path to the model tarball) and the Docker image for inference. This resource defines the model that SageMaker will use for hosting. AWS::SageMaker::Endpoint is incorrect because it defines the live inference endpoint, which hosts a model, but doesn't define the model's properties. AWS::SageMaker::NotebookInstance is incorrect as it provisions a development environment, not an ML model for deployment. AWS::SageMaker::Pipeline is incorrect because it orchestrates ML workflows, not individual models.