Using Azure AI Foundry to manage all your Large Language Models
summary
Live demos show Azure AI Foundry deploying and managing real models, not slides about architecture: a model catalog with >1,800 models (including GPT40), hub-and-project tenancy, RBAC, and playground-based testing. Deployment details include quotas like 8,000 tokens per minute, SDK examples for Python/C#, and marketplace-style benchmarking. Playground chats respond correctly and system messages enforce behavior; RAG integration uses Azure AI Search and blob/Cosmos connectors. One file upload hiccup appears in the demo but core deploys and tests work end to end.
AI-written summary · about
chapters (10)
- 0:00Introduction and Session Overview
- 1:48Future of Work with AI
- 5:10Azure AI Foundry Overview
- 10:51Live Demo: Azure AI Foundry Setup
- 17:42Azure AI SDK for Developers
- 18:58Deploying Large Language Models
- 31:32Integrating Traditional AI Services
- 39:37RAG Architecture: Enhancing AI with Custom Data
- 45:40Responsible AI Framework
- 47:00Conclusion and Final Thoughts
related talks
48mDeploy large language models responsibly with Azure AI
59mLearn Live: Introduction to Azure OpenAI Service
38mSafe Evaluation and Rollout of AI Models
46mMultimodel Multimodal and Multiagent innovation with Azure AI
22mQuickly deploy NVIDIA optimized generative AI on Azure AI Studio
47mAccelerate your AI journey with Azure AI model catalog
20mUnlocking Azure OpenAI: A 20-Minute Introduction
39mDisrupt and Grow with Microsoft Azure AI