HP intros ZGX Nano G1n AI workstation: powered by NVIDIA's new GB10 Superchip

HP has just unveiled its new compact AI workstation with the introduction of the new ZGX Nano G1n system, powered by NVIDIA's powerful new GB10 Superchip.

2

The new HP ZGX Nano G1n compact AI workstation measures just 15 cm x 15 cm x 5 cm, making for a tiny system with huge computing power courtesy of the powerful NVIDIA GB10 Superchip, delivering both powerful Grace CPU cores and a Blackwell-based GPU.

HP does something a little special with its ZGX Nano G1n workstation system with its ZGX Toolkit, which is a software stack that runs on NVIDIA's AI Stack (DGX OS) that provides a smoother development experience. HP's new mini AI workstation packs 1000 TOPS of AI compute power thanks to the NVIDIA Blackwell GPU, connected to the 20-core NVIDIA Grace CPU through NVLink-C2C chip-to-chip interconnect.

HP's new ZGX Nano G1n mini AI workstation also features up to 128GB of unified LPDDR5X memory that gets shared between the CPU and GPU, removing any memory bottlenecks. There's also scalability through the interconnect, allowing users to link two units together through a 200Gbps QSFP/ConnectX-7 interface that instantly doubles the performance, perfect for larger AI models.

On the connectivity side, we've got high-bandwidth ports including USB Type-C 20Gbps and super-fast 10GbE ethernet, with Wi-Fi 7 for fantastic wireless speeds as well as Bluetooth 5.4 connectivity. There's a single HDMI port on the HP ZGX Nano G1n compact AI workstation, but we don't know which version of HDMI... let's hope it's the faster HDMI 2.1 standard.

HP ZGX Nano G1n compact AI workstation details:

• NVIDIA DGX OS and AI Software Stack: Hit the ground prototyping, fine-tuning, and inferencing with familiar NVIDIA DGX; OS and NVIDIA; AI software stack-purpose-built for AI development.
• NVIDIA GB10 Grace Blackwell Superchip: Get up to 1000 TOPS of AI compute with an NVIDIA Blackwell GPU. Supercharge data preprocessing and orchestration with a Grace 20-core Arm CPU.
• Unified System Memory: Run AI development and testing workloads with AI models of up to 200 billion parameters at your desk with 128 GB of coherent unified system memory.
• Fast, Secure Storage: Choose between 1 or 4 TB of NVMe M.2 self-encrypted storage to work efficiently with large files and keep more of your data local and secure.

HP's new ZGX Nano G1n AI Station will launch later this year, with pricing still to be decided.