Quiz 7 Question 18 of 20

A machine learning team is choosing among several candidate language models in the Azure AI Foundry model catalog for a new customer-facing assistant. Before committing to one, they want to compare quality, cost, and latency metrics for multiple models side by side using the same benchmark prompts, without deploying every candidate to production first.

Select an answer to reveal the explanation.

Motivation