NVIDIA Brings Local AI Agents To Its Most Powerful Workstation PC, The DGX Station, With The NVIDIA Agent Toolkit & Omniverse

NVIDIA has enabled DGX Station users to run personal AI agents locally through its NVIDIA Agent Toolkit, which can be set up in just three steps.

DGX Station With GB300 Is A Super Workstation PC, & You Can Now Easily Run Super AI Agents On It With NVIDIA's Agent Toolkit

Agentic AI has brought new kinds of use cases for PCs, especially powerful ones such as NVIDIA's DGX Station, which packs the GB300 "Blackwell Ultra" GPU. To make full use of its power, NVIDIA is offering its Agent Toolkit software, which unleashes these capabilities in just 30 minutes (the time it takes to set up the software).

The NVIDIA Agent Toolkit is a complete software stack that brings NVIDIA NemoClaw, Neomotron 3 Ultra, and Omniverse together. With NVIDIA Omniverse libraries, agents can access tools and skills in a secure runtime locally without needing to connect to the internet.

Developers can also connect multiple systems to serve concurrent users, more agents, and bigger model sizes.

• NVIDIA NemoClaw offers open blueprints for building custom autonomous agents, packaging the model, harness, and runtime together as a starting point for teams building specialized, domain-specific agents.
• NVIDIA Nemotron 3 Ultra, a frontier 550-billion-parameter open model, is optimized to run on DGX Station GB300 systems and serves as the model layer that teams can customize for their own domains.
• NVIDIA Omniverse libraries extend agent skills into physics simulation and 3D asset workflows, giving creative and engineering professionals tools that go well beyond general-purpose agent capabilities.
• NVIDIA OpenShell, the open-source secure runtime, keeps agents sandboxed and governed according to defined policies for how agents interact with tools, systems and data.
• NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip delivers data-center-level performance from the desk on DGX Station, with up to 20 petaflops of FP4 AI compute and 748GB of coherent memory to run large models such as Nemotron Ultra.
• NVIDIA ConnectX-8 SuperNIC delivers up to 800GB/s of bandwidth in DGX Station, delivering extremely fast, efficient network connectivity, and supports linking up to two DGX Stations to further scale model capacity and performance.

With the Neomotron 3 Ultra "Open" model, teams that are running agents at scale can utilize the hardware's leading-edge performance without worrying about token cost, since the hardware purchase is the only thing they will be paying for.

Besides that, NVIDIA has announced a blueprint for integrating Omniverse into libraries, which can enable devs to prepare 3D scenes for Physical AI workflows through NemoClaw's "RTX Sensor Simulation". And most importantly, teams running OpenClaw can now make use of Neomotron 3 Ultra and Omniverse tools for local inferencing on DGX Station.

NVIDIA is also offering two new playbooks to help developers build and run agents with NemoClaw and dual-node deployments:

• "Connect Two DGX Stations for Distributed Workloads” guides users on how to run powerful models with scalable performance by connecting up to two DGX Station systems.
• “Run NemoClaw With a Local LLM / On Dual DGX Station” provides a walkthrough on how to set up NemoClaw on DGX Station, enabling out-of-the-box agents with frontier intelligence with Nemotron Ultra.

NVIDIA is also working with its partners such as Adobe, Blender, Unreal, Unreal Engine, Epic Games, SideFX, Foundry, and Canva to integrate MCP (Model Context Protocol) connections, allowing AI agents to work inside the tools where scenes, shots, timelines, assets, and edits are being rendered.

With that said, NVIDIA is expanding its local AI capabilities with powerful tools and libraries, now available on its most powerful workstation PC platform, the DGX Station. The DGX Station is available for order through ASUS, Dell, Exxact, Gigabyte, HP, MSI, and Supermicro.

Follow Wccftech on Google to get more of our news coverage in your feeds.