NVIDIA’s Open Models on Microsoft Foundry:

2026-03-16 · Source: Microsoft Foundry Blog articles · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Cloud Computing & IT Infrastructure, Robotics & Autonomous Systems · Depth: Intermediate, short

Summary

Microsoft and NVIDIA are expanding the Microsoft Foundry platform through deeper integrations, bringing NVIDIA's open models, accelerated computing, and AI development frameworks into the Foundry ecosystem. This collaboration, announced at GTC 2026, focuses on specialized agentic systems powered by Nemotron models, sovereign and on-prem AI deployments via Foundry Local, and production-grade physical AI workflows built on Azure and NVIDIA AI platforms. Developers can now build and deploy specialized AI agents using NVIDIA Nemotron models directly within Foundry, leveraging Azure infrastructure accelerated by NVIDIA GPUs. The platform will offer Nemotron Nano 9B v2, Llama 3.1 Nemotron Nano VL 8B, Llama 3.3 Nemotron Super 49B v1.5, and NVIDIA Nemotron Super 3, with future expansion to include Nemotron 3 family models, Nemotron Speech, Nemotron Vision, and AI Safety models. Foundry Local extends capabilities for sovereign environments, supporting NVIDIA RTX PRO 6000 Blackwell Server Edition and future NVIDIA Rubin platforms. An open Azure Physical AI toolchain integrates NVIDIA blueprints with Azure services for robotics and physical AI systems.

Key takeaway

For AI Architects and CTOs evaluating enterprise AI platforms, this expanded Microsoft Foundry and NVIDIA integration offers a compelling solution for unified AI development and deployment. You can now build specialized AI systems once and deploy them consistently across cloud, hybrid, and sovereign environments, addressing data sovereignty and operational complexity challenges. Consider leveraging Nemotron models and Foundry Local for your next-generation agentic or physical AI initiatives.

Key insights

Microsoft Foundry integrates NVIDIA's open models and accelerated computing to unify AI development and deployment across diverse environments.

Principles

• Unified platforms reduce AI fragmentation.
• Open-weight models enable customization.
• Sovereignty requires local AI deployment.

Method

Microsoft Foundry provides a managed compute environment for deploying and customizing NVIDIA Nemotron models on Azure, supporting specialized agentic, sovereign, and physical AI systems.

In practice

• Deploy Nemotron models via NVIDIA NIM microservices.
• Utilize Foundry Local for sovereign AI deployments.
• Explore Azure Physical AI toolchain for robotics.

Topics

• Microsoft Foundry
• NVIDIA Nemotron
• Agentic AI
• Sovereign AI
• Physical AI

Best for: CTO, AI Architect, Computer Vision Engineer, AI Engineer, Machine Learning Engineer, MLOps Engineer

Related on AIssential

• Nvidia Launches Physical AI Models for Robots — Nvidia's new physical AI models and frameworks aim to democratize intelligent robot development and deployment.
• NVIDIA and ServiceNow Partner on New Autonomous AI Agents for Enterprises — Enterprise AI agents require robust governance, secure execution, and domain-specific adaptability for scalable deployment.
• GTC: What are NVIDIA's New Open Models ? — NVIDIA's open model expansion targets agentic, physical, and biomedical AI, moving beyond language-centric systems.
• Nemotron Labs: What OpenClaw Agents Mean for Every Organization — Autonomous AI agents running persistently and locally offer significant advantages for continuous, high-iteration tasks.
• NVIDIA brings agents to life with DGX Spark and Reachy Mini — NVIDIA's DGX Spark and Reachy Mini enable building private, customizable, real-world AI agents using open models.
• Why Nvidia builds open models with Bryan Catanzaro — NVIDIA's Nemotron project strategically advances AI infrastructure through open models, data, and research, driven by both internal product needs and ecosystem growth.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Microsoft Foundry Blog articles.