Other Published:
NVIDIA and Microsoft bring Nemotron 3.5 and RTX PRO 6000 Blackwell support to Azure Local

What happened
At Microsoft Ignite, NVIDIA and Microsoft announced on November 19 that Foundry Local is expanding support for NVIDIA's Nemotron 3.5 open models, including retrieval models, in Azure Local environments. Azure Local will support NVIDIA RTX PRO 6000 Blackwell GPUs from hardware partners and is expanding to the Jetson Thor and Jetson Orin edge platforms. In the cloud, Nemotron 3.5 Super and Lightning are offered on Foundry Managed Compute, with Bolt and Transcribe listed as coming soon. NVIDIA also cites EY, which adapted Nemotron models to classify purchasing records and reports a 55% accuracy improvement on its evaluation dataset at 15x lower cost than proprietary frontier models.
Why it matters locally: Companies that already manage their own servers with Microsoft tools get an officially supported way to run open models on NVIDIA workstation GPUs outside the cloud.
Three ways to read it
For the market
This looks like Microsoft and NVIDIA treating on-premises AI as a standard enterprise product rather than a do-it-yourself project, which could pull cautious IT departments toward local deployment. The EY figures are the customer's own results on its own data, so read them as one case rather than a general benchmark.
In plain words
Microsoft's system for running company servers in your own building will be able to run NVIDIA's open AI models on NVIDIA's professional graphics cards. For a business, that means AI that works on internal documents without the documents leaving the premises.
For practitioners
The RTX PRO 6000 Blackwell has 96 GB of memory per card in our data, which should leave room for a chat model plus embedding and rerank models on one GPU, so a retrieval stack could likely sit on a single node. NVIDIA gives no dates or supported server list for Azure Local, so confirm hardware partner availability and which Nemotron 3.5 variants Foundry Local actually ships before planning a rollout.
Sprout's insight
Sprout compares this with other brands and with where real applications are heading. Free for members.
Sign in free to read it →Subscribe to Sprout's weekly
One email a week: the top stories and a taste of Sprout’s insights. Unsubscribe any time.
Source
Read the official source: NVIDIA Blog: NVIDIA and Microsoft Expand Accelerated Computing From Azure to Windows PCs ↗Summaries are written by OnPrem World from the official source. We link to the original and never republish it.
On this site
More news
- Microsoft publishes Aion 1.0 Instruct, the 2.3B model that becomes Windows' built-in AI by end of January 2027
- RTX Spark compact desktops from ASUS, Dell and MSI go on sale with up to 128 GB of unified memory
- NVIDIA reports $115.8 billion quarter; highlights include a deal to acquire Hugging Face and a 64GB DGX Spark
- Open WebUI v0.13.4 moves personal connections into Tether, sandboxes agents on your own computer and runs local MCP servers
- vLLM v0.34 sizes the KV cache from real free memory, removes Model Runner V1 and adds Mistral Large 4