Microsoft will offer Nvidia's models and tooling in Microsoft Foundry.

The claim is that Microsoft will offer Nvidia's models and Nvidia's tooling in Microsoft Foundry (Azure AI Foundry). No deadline is stated, so judge it as of the evaluation date. True: Microsoft documentation, the Foundry model catalog or a dated announcement shows that Nvidia models (e.g. Nemotron or other Nvidia-published models) and Nvidia tooling (e.g. NIM microservices, NeMo or other Nvidia developer tools) are available in Foundry after the June 2026 statement. Earlier availability that continues also counts. Mostly true: both are available, with a minor gap in some detail, such as the tooling being in preview only. Mostly false: only one of the two (models or tooling) is available, or the offering exists only as an announced plan. False: positive evidence that neither is offered, such as a dated catalog or documentation review showing no Nvidia models or tooling in Foundry, or a statement that the plan was dropped.

Set before the research began.

1. E1 Deploy Microsoft Foundry Models to managed compute with pay-as-you-go billing (classic) - Microsoft Foundry (classic) portal Microsoft Learn · 20 August 2026

   “NVIDIA inference microservices (NIM) are containers that NVIDIA builds for optimized pretrained and customized AI models serving on NVIDIA GPUs. You can deploy NVIDIA NIMs available on Foundry model catalog with a Standard subscription to the NVIDIA NIM SaaS offer on Azure Marketplace.”

2. E2 Foundry Models | Microsoft Azure Microsoft Azure · 21 October 2026

   “Microsoft Foundry offers a range of flagship models—including Azure OpenAI, Anthropic Claude, Meta, Mistral AI, DeepSeek, xAI, Cohere, HuggingFace, NVIDIA, Fireworks AI and more—available through serverless pay-as-you-go or managed compute offerings.”

3. E3 Azure# NVIDIA · 4 November 2026

   “Microsoft Foundry: Managed deployment using Azure ML Managed Online Endpoints.”

4. E4 NVIDIA-Nemotron-3-Super-NI... | Model Catalog | Microsoft Foundry Microsoft Foundry · 6 June 2026

   “Nemotron-3-Super-120B-A12B is a large language model (LLM) trained by NVIDIA, designed to deliver strong agentic, reasoning, and conversational capabilities.”

The claim resolves true: as of today, both Nvidia models and Nvidia tooling are offered in Microsoft Foundry. The Foundry model catalog lists NVIDIA's Nemotron-3-Super model as a NIM microservice (E4, dated 2026-06-06), and Microsoft's Foundry Models product page names NVIDIA among its model providers (E2). Microsoft's documentation says NVIDIA NIM containers can be deployed from the Foundry model catalog (E1), and NVIDIA's own NIM documentation lists Microsoft Foundry as a managed deployment option (E3). E3 is a source independent of Microsoft.