Local AI development with Foundry Local and .NET Aspire | DEM520

When we start building apps infused with AI it begs the question of how we tackle the local dev experience. Do we provision cloud resources, if so, how do we secure them, manage rate limits and cost? Or do run local models, and how do we set them up? What about vector databases, data ingestion, and all those auxiliary systems? It all sounds complicated, that is until we start leveraging the power of .NET Aspire and see that we can have it as easy as clone repo and dotnet run! 𝗦𝗽𝗲𝗮𝗸𝗲𝗿𝘀: * Aaron Powell 𝗦𝗲𝘀𝘀𝗶𝗼𝗻 𝗜𝗻𝗳𝗼𝗿𝗺𝗮𝘁𝗶𝗼𝗻: This is one of many sessions from the Microsoft Build 2025 event. View even more sessions on-demand and learn about Microsoft Build at https://build.microsoft.com DEM520 | English (US) | AI, Copilot & Agents Related Sessions: BRK146 -- https://build.microsoft.com/sessions/BRK146?wt.mc_id=yt_PLlrxD0HtieHg4OIwuQGsVgpQhTKl4fFdE #MSBuild Chapters: 0:00 - Challenges of Running AI Locally vs. Cloud 00:02:58 - Demonstration of Foundry Local Usage 00:03:37 - Distributed Application Concept Explanation 00:04:02 - Model Download and Service Management with Aspire 00:05:50 - Configuration of Foundry Local Resource 00:06:33 - Explanation of the Chat Model Requirement and Initial Mention of Application's Start 00:10:47 - Options for downloading models and continuation plans 00:11:50 - Explanation of automatic model version and device compatibility selection 00:12:11 - Details on CPU model download 👩‍💻👨‍💻 Great devs don’t build alone. ⏱️ In a fast-paced dev world, there’s no time to hunt for help. That’s why we created the Azure AI Foundry Developer Community — your one-stop hub for: 🌐 A vibrant forum for Q&A + code 🎙️ DevBlogs with the latest updates 🎮 Real-time Discord chats, events & learning 🔗 Join the Forum -- https://github.com/orgs/azure-ai-foundry/discussions 🔗 Hop on Discord -- https://discord.com/invite/azureaifoundry