Microsoft and Hugging Face expand collaboration
Microsoft and Hugging Face expand collaboration to make open models easy to use on Azure
Today at the Microsoft Build conference, Satya Nadella announced an expanded collaboration with Hugging Face, to make its wide diversity of open models easy to deploy on Azure secure infrastructure.
If you head over to Azure AI Foundry today, you will find a vastly expanded collection of 10,000+ Hugging Face models you can deploy in a couple clicks to power AI applications working with text, audio and images. And we’re just getting started!
It’s time to build - an expanded collaboration
2 years ago, Microsoft and Hugging Face started a collaboration to make open models more easily accessible on Azure - back then the Hub was home to 200,000 open models.
With now close to 2 million open models on Hugging Face, covering a wide diversity of tasks, modalities, domains and languages, it was time to take our collaboration to the next level. The new collaboration announced today creates a framework for mutual success to vastly expand how Azure customers can benefit from Hugging Face.
"Open source is how AI moves faster—with transparency, choice, and community at its core. This collaboration represents our commitment to that momentum. By combining Hugging Face’s vibrant model ecosystem with Azure’s secure, enterprise-grade infrastructure, we’re giving developers the freedom to pick the best model for the job—and helping organizations innovate safely and at scale.”
-- Asha Sharma, Corporate Vice President at Microsoft
Making more open models easily accessible to Azure customers, for secure deployment alongside company private data, will enable enterprises to build AI applications and agents while being fully in control of their technology and data.
We’re enabling companies to take control of their AI destiny, deploying the best open models securely within their Azure account, to build AI applications they can trust and verify.
-- Clement Delangue, CEO and cofounder at Hugging Face
How to use Hugging Face in Azure AI Foundry
Let’s head over to Azure AI Foundry, and select the Model Catalog. Here you can now find over 10,000 models under the Hugging Face Collection.
To make the Hugging Face Collection on Azure AI Foundry enterprise-ready, we are only featuring models:
having successfully passed Hugging Face security tests to screen for any vulnerability, including with ProtectAI Guardian and JFrog security scanner
with model weights stored in safetensors format, avoiding potential Pickle vulnerabilities
without remote code, to avoid any arbitrary code insertion at runtime.
In addition, Microsoft and Hugging Face will continuously test inference containers for vulnerabilities to maintain and patch them as needed.
Now let’s say you want to deploy for instance the popular Microsoft Phi-4 Reasoning Plus open model.
First, let’s select the model in the Hugging Face Collection in Azure AI Foundry, and click the “Deploy button”. The form allows you to select a VM, instance count and deployment parameters, and start the deployment process with just another click!
Now, if you prefer browsing models on the Hub, you can also start from the model page - the “Deploy on Azure ML” option will take you to the same deployment option within Azure AI Machine Learning Studio.
More Hugging Face to come in Azure AI Foundry
We are really excited about all the new Hugging Face models and modalities now directly available within Azure AI Foundry, but we’re not going to stop there!
In the weeks and months to come, you can expect a rolling thunder of updates:
Day-0 releases - Hugging Face will collaborate with Microsoft to make new models from top model providers available in Azure AI Foundry the same day they land on Hugging Face
Trending models updates - Hugging Face will continuously monitor trending models to enable them on Azure AI Foundry on a daily basis
New modalities - Hugging Face and Microsoft will work together to enable more modalities and domain-specific tasks, including video, 3D, time series, protein and more.
Agents and tools - Small, efficient, specialized open models make them ideal to build powerful but secure AI agents and applications
If you’re on Azure, it’s time to build with open models!
readers loved this