Open-source models fuel modern AI, but running them in production is hard. In this lightning talk, Microsoft and Hugging Face demonstrate how to deploy and scale Hugging Face models using Foundry Managed Compute inside Azure AI Foundry. Through a fast, end-to-end demo, developers will see how to go from model discovery to production inference without managing GPUs, while gaining autoscaling, governance, and enterprise-grade performance.
If you're following these steps at your own pace:
- Clone this repository
- Set up your development environment
By the end of this demo, you will be able to:
- Discover and evaluate Hugging Face open-source models using the Azure AI Foundry model catalog.
- Deploy a Hugging Face model to production using Foundry Managed Compute, without managing GPUs directly.
- Understand how Azure AI Foundry provides autoscaling, governance, and enterprise-grade performance for open-source model inference.
Try these prompts with GitHub Copilot to explore the topics from this demo. Open Copilot Chat in Visual Studio Code (Ctrl+Alt+I on Windows/Linux, Cmd+Shift+I on Mac), paste a prompt, and see what you learn. Try connecting the Microsoft Learn MCP Server for the latest official documentation.
Use these as a starting point — or write your own!
Prompts coming soon — check back after the session content is finalized.
- Microsoft Foundry — the unified Azure platform for enterprise AI operations, covering agents, 1,900+ models, SDKs, and the Foundry portal.
- Hugging Face models in Microsoft Foundry — deploy any Hugging Face Hub model to a managed compute endpoint in Microsoft Foundry for scalable, secure, real-time inference.
- Foundry Managed Compute — deploy and serve catalog models (including Hugging Face open-source models) using managed online endpoints with autoscaling and enterprise-grade performance.
| Resource | Description |
|---|---|
| https://aka.ms/build26-next-steps | Explore lab and session repos to further your learning from Microsoft Build |
| https://aka.ms/hf/docs/microsoft-azure | Hugging Face documentation for deploying models on Microsoft Azure |
| https://aka.ms/hf/docs/microsoft-azure/one-click-deploy | One-click deploy Hugging Face models to Microsoft Azure |
| https://aka.ms/hf/foundry-models | Hugging Face models available in Microsoft Foundry |
The Microsoft Learn MCP Server gives your AI agent direct access to Microsoft's official documentation — grounded, up-to-date answers about the products and services covered in this demo.
VS Code — One click installation:
GitHub Copilot CLI — Run this to install the Learn MCP Server as a plugin:
/plugin install microsoftdocs/mcp
For more info, other clients, and to post questions, visit the Learn MCP Server repo.
 Osi Otugo 📢 |
 Vaidyaraman Sambasivam 📢 |
 Jeff Boudier 📢 |
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit Contributor License Agreements.
When you submit a pull request, a CLA bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., status check, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.
This project may contain trademarks or logos for projects, products, or services. Authorized use of Microsoft trademarks or logos is subject to and must follow Microsoft's Trademark & Brand Guidelines. Use of Microsoft trademarks or logos in modified versions of this project must not cause confusion or imply Microsoft sponsorship. Any use of third-party trademarks or logos are subject to those third-party's policies.