Where do you register a model for governed use?
Why C is the answer
Unity Catalog, using MLflow
Registering to Unity Catalog through MLflow gives a model versions, permissions and lineage.
In detail
- MLflow registers the model to Unity Catalog.
- That gives versions and governed access.
- Use three-level names, such as catalog, schema and model.
Use case
An illustrative scenario, not a real incident.
A team registers each approved chain in Unity Catalog through MLflow. Others can find it, permissions apply, and versions stay traceable.
See it in different ways
flowchart TD A["RAG model"] --> B["Model flavor"] A --> C["Embedding model"] A --> D["Retriever"] A --> E["Dependencies"] A --> F["Input example"] A --> G["Model signature"]
sequenceDiagram participant C as pyfunc chain participant M as MLflow participant U as Unity Catalog participant S as Model Serving C->>M: Log the model M->>U: Register a version U->>S: Deploy to an endpoint
flowchart LR S0["Log with MLflow"] --> S1["Register to Unity Catalog"] --> S2["Grant access"] --> S3["Deploy a version"]
Key point: Register models in Unity Catalog with MLflow.
Why the other options are wrong
- A. The workspace file system
- The workspace file system does not govern model versions.
- B. A shared drive
- A shared drive has no access model for models.
- D. A local notebook folder
- A local folder has no versions or governance.