Question

A retail company's ML team has selected an open-source foundation model from Vertex AI Model Garden after evaluating several options. They now want to make this model available to their internal applications as a scalable, managed prediction service with a stable HTTPS address, without provisioning and maintaining their own servers. What is the most appropriate next step within Vertex AI?

Accepted Answer

Deploy the model to a Vertex AI endpoint to serve online predictions through managed, autoscaling infrastructure

Answer

Copy the model weights to a self-managed Compute Engine GPU cluster and expose a custom API

Answer

Call the pre-built Natural Language API instead, since it already provides a hosted endpoint

Answer

Use the Gemini app so business users can query the model through a chat interface