You manage an Azure Al Foundry project.

You plan 10 build a RAG solution. The solution must include two models:

• One for text output, named Model1. This model must resemble human language and read naturally.

• One for creating embeddings, named Model2. This model must maximize the retrieval of relevant results (high recall)

You need to compare different models by using benchmarking metrics to select the appropriate models for Model1 and Model?