You can deploy LLMs in Azure AI Foundry using either Azure OpenAI models such as GPT-4, GPT-4o or open-source models like Llama 3, Phi-3, Mistral etc.

Some steps below:

1. Prerequisites

   • Have an Azure subscription: https://azure.microsoft.com/free

   • Request Azure OpenAI access: https://aka.ms/oai/access

2. Create a Hub

   • Go to https://ai.azure.com → Get started → Create a hub.

   • Choose a name, subscription, resource group, and region (e.g. East US).

   • Click Review + Create → Create.

3. Create a Project

   • Inside the hub, click + New project → name it (e.g. llm-demo) → Create.

4. Deploy a Model

   • Option A: Azure OpenAI model

     • Go to Models + endpoints → Azure OpenAI → + Deploy model.

     • Choose a model (e.g. gpt-4o-mini) and click Deploy.

   • Option B: Open-source model

     • Go to Model catalog, pick something like Llama 3 or Phi-3, click Deploy, choose compute, and wait for deployment.

5. Test in Playground

   • Open Playground, select your model deployment, and run prompts interactively.

If you want, I could show you how to use your deployed models for downstream tasks???