Question

A GenAI engineer has registered a fine-tuned model in Unity Catalog at `ml.prod.support_summarizer` and deployed it to a Model Serving endpoint. The endpoint runs under a service principal used by the production application. During testing, queries to the endpoint fail with a permission error when the model version is loaded. The engineer confirms the endpoint configuration is otherwise correct. What is the MOST appropriate action to resolve this while following least-privilege governance practices?

Accepted Answer

Grant the service principal `EXECUTE` on the model and `USE CATALOG`/`USE SCHEMA` on the parent catalog and schema in Unity Catalog

Answer

Grant the service principal `ALL PRIVILEGES` on the `ml.prod` catalog so it can access any current and future models

Answer

Add the service principal as a workspace admin so it inherits access to all registered models

Answer

Set the model's storage location ACLs to public-read so the serving endpoint can load the artifacts