Which AWS capability captures metadata and details about ML models and instance data to support governance and reporting?

Amazon SageMaker Model Cards is the correct answer because it provides a centralized, human-readable record of ML model details, including metadata, training data, evaluation metrics, and intended uses. This capability is specifically designed to support model governance, auditing, and reporting requirements by documenting key aspects of the model lifecycle. Amazon SageMaker Debugger helps analyze and debug training jobs by capturing and analyzing tensors during training, not for general model metadata. Amazon SageMaker Model Monitor continuously monitors the performance of deployed models for drift and anomalies, focusing on runtime behavior rather than static model details. Amazon SageMaker JumpStart offers pre-built solutions and models for quick deployment, but it doesn't serve as a repository for governance-related model metadata.