An ML engineer created a binary classification model outside SageMaker and stored the model artifacts in S3. The engineer and a SageMaker Canvas user are in the same SageMaker domain. What requirements must be met so the engineer can share the model with the Canvas user? (Choose two.)

For a SageMaker Canvas user to utilize a model created outside SageMaker, two key requirements must be met. First, the model must be registered in the SageMaker Model Registry. The Model Registry acts as a central catalog for managing models, making them discoverable and shareable within SageMaker, including for Canvas. Second, the Canvas user (or the IAM role assumed by Canvas on their behalf) needs appropriate permissions to access the S3 bucket where the model artifacts are stored. Without S3 access, Canvas cannot retrieve the model files. Incorrect options: The ML engineer and the Canvas user must belong to separate SageMaker domains: This is incorrect; they can be in the same domain, as stated in the prompt. The ML engineer must list the model on AWS Marketplace: Listing on AWS Marketplace is for commercial distribution, not for internal sharing within a SageMaker domain. The ML engineer must deploy the model to a SageMaker endpoint: While deploying to an endpoint is common for inference, Canvas can use models directly from the Model Registry without requiring a pre-existing endpoint.