Hugging Face Models on Foundry Managed Compute

Microsoft's Foundry platform now offers a curated catalog of open-weight models from the Hugging Face ecosystem, deployable with a single click. This integration, called Foundry Managed Compute, provides a managed GPU service that handles all the complex infrastructure, security patching, and runtime optimization. Each model undergoes a rigorous curation process where Microsoft screens it for security, reviews licenses, and prepares it for enterprise environments on runtimes like vLLM and SGLang. Developers can access this weekly-refreshed library of state-of-the-art models through the same secure endpoint and SDKs used for proprietary models, simplifying the use of open-source AI.