NVIDIA Corp. (NASDAQ: NVDA) announced the DGX Station for Windows, a desktop AI supercomputer designed to run AI models with up to 1 trillion parameters locally. The system is built on the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip and is scheduled for availability in the fourth quarter of this year.

The announcement was made at NVIDIA GTC Taipei. The DGX Station for Windows represents the first desktop AI supercomputer to bring NVIDIA GB300 Grace Blackwell-class AI infrastructure into the Windows ecosystem, according to the company's press release.

The system features up to 748GB of coherent memory and up to 20 petaflops of FP4 performance. It can be paired with an NVIDIA RTX PRO 6000 Blackwell Workstation GPU for additional computing capabilities. The device includes the NVIDIA ConnectX-8 SuperNIC, which supports networking speeds up to 800Gb/s.

"As enterprises scale AI agents across their organizations, they need AI infrastructure that can connect directly to the applications and workflows that power their business," said Chris Marriott, vice president of enterprise platforms at NVIDIA.

Microsoft collaborated on the development of the system. "Today, we're taking that collaboration to the next level, scaling the full power of Windows from thin-and-light PCs to data-center-class workstations with DGX Station powered by GB300," said Pavan Davuluri, executive vice president of Windows + Devices at Microsoft.

The DGX Station for Windows will support NVIDIA OpenShell on Windows, which creates isolated sandboxes for individual agents and enforces security policies at the system level.

The system will be available from manufacturers including ASUS, Dell Technologies, GIGABYTE, HP, MSI and Supermicro in the fourth quarter of this year.