Azure AI Foundry PM, Microsoft Alexander Hughes Pamela Fox Principal Cloud Advocate, Microsoft Spotlight This Week: Open Source Models Build Your Model IQ https://aka.ms/model-mondays 

Senior AI Advocate Nitya Narasimhan, PhD Sharmila Chockalingam Senior Product Marketing Mgr. Welcome to S1:E07 Model Mondays Build Your Model IQ https://aka.ms/model-mondays 

Model Mondays Highlights 5 Things To Know This Week Sharmila Chockalingam 

❶ · Try Command A & Embed 4 https://aka.ms/model-mondays Agentic RAG Deploy high-performance LLMs and semantic search engines directly into production with Command A and Embed 4 on Azure AI Foundry. 

❷ · Try MAI-DS-R1 https://aka.ms/model-mondays MAI-DS-R1 A new open-weights DeepSeek R1 model variant with improved responsiveness on blocked topics while maintaining its reasoning capabilities. 

❸ · Try OpenAI Models https://aka.ms/model-mondays GPT 4.1 and o Series GPT 4.1 models represent something special—a generational advancement in coding, reasoning, and understanding context o3 — Their most advanced model yet, strong in coding, math, science, and visual tasks o4-mini — Smaller, faster, and remarkably capable in technical domains, at lower cost. 

❹ · Try GH Copilot https://aka.ms/model-mondays Github Copilot New models from the last 7d in GitHub Copilot. 

❺ · Try Azure MCP Server https://aka.ms/model-mondays Azure MCP Server Connect AI agents to Azure services like Storage, Cosmos DB, AppConfig, Monitor and more - with natural language. 

Welcoming the Llama 4 Herd to Azure AI Foundry What is Llama 4? What are the model's capabilities and advantages? How can you get started with it? Alexander Hughes Product Manager II – Azure AI Foundry Models 

❶· What is Llama 4? Llama 4 herd of state-of-the- art natively multimodal models on Azure AI Foundry We’re sharing the first models in the Llama 4 herd, which will enable people to build more personalized multimodal experiences. According to Meta, Llama 4 Scout is one of the best multimodal models in its class and is more powerful than Meta’s Llama 3 models, while fitting in a single H100 GPU. 

❸· What is MoE architecture? Mixture of Experts (MoE) Architecture achiev es strong performance without sacrificing compute cost Llama models are fully open source, which allows for their cross-industry benchmarking and use What makes Llama 4 special? 

ai.azure.com/explore/models/Llama-4-Scout-17B-16E/version/1/registry/azureml-meta ❹· How can I use Llama 4? How to get started with Llama 4 How to use inferencing code in your application To get started: 1. Navigate to Azure AI Foundry or GitHub Models 2. Search for Llama 4 Scout or Llama 4 Maverick in the model catalog 3. In Foundry, select 'Deploy' and then the 'Serverless API' option. In GitHub, search for the model and test it in the playground. You can select 'Use this model' to view the code you can use in your application. 

Learn about open source like Llama models and how to deploy the latest Llama4 model and its uses cases. Hands on with Open-Source Models Pamela Fox, Principal Cloud Advocate - Microsoft Models 

Hands-on with Forecasting Models Learn about the Nixtla TimeGEN-1 model for time-series forecasting with applications to finance, retail & more. Khuyen Tran, Developer Advocate - Nixtla Next Week 

Join the #model-mondays channel https://aka.ms/model-Mondays/chat Q&A Office hours every Friday on Discord – bring questions, share your projects! 

Watch the Livestream Join the Conversation Explore the Resources Mon 1:30pm ET Microsoft Reactor https://aka.ms/model-mondays/RSVP Fri 1:30pm ET Azure AI Discord https://aka.ms/model-mondays/Discord Watch or Star GitHub Repo https://aka.ms/model-monday #ModelMondays 

Build Your Model IQ https://aka.ms/model-mondays Model Mondays 10:30 am PT | 1:30 pm ET