Limited Time Offer: Save $350 on your Ai Cluster → Shop Now

--

Days

--

Hours

--

Mins

--

Secs

Sale Ended

HP ZGX Nano G1n AI Station – NVIDIA GB10, 128GB, 4TB SSD
HP ZGX Nano G1n AI Station – NVIDIA GB10, 128GB, 4TB SSD
HP ZGX Nano G1n AI Station – NVIDIA GB10, 128GB, 4TB SSD
HP ZGX Nano G1n AI Station – NVIDIA GB10, 128GB, 4TB SSD

Vendor:HP

HP ZGX Nano G1n AI Station – NVIDIA GB10, 128GB, 4TB SSD

$6,999.00

The HP ZGX Nano G1n AI Station brings NVIDIA GB10 Grace Blackwell AI compute, 128GB of unified memory, and 4TB of NVMe storage to a compact desktop system. Built for local prototyping, fine-tuning, inference, and edge AI development, it pairs NVIDIA DGX OS with HP ZGX Toolkit to help move AI workflows from experimentation to deployment without relying on cloud resources.

Configuration: 1 HP ZGX Nano G1n

In stock, ready to ship
Product datasheet

Pairs well with

30-Day AISuperStore Guarantee 

TECHNOLOGY AT A GLANCE

Explore the capabilities

Local AI compute

Prototype, fine-tune, and run inference locally with NVIDIA GB10 Grace Blackwell AI performance.

128GB unified memory

Work with large datasets and compatible AI models using 128GB of coherent LPDDR5X unified memory.

200G networking

Two 200G QSFP ports powered by NVIDIA ConnectX-7 enable high-speed connectivity to compatible systems and networking.

Compact AI station

A 5.9-inch-square desktop system designed to bring serious AI capability to the desk, lab, or edge deployment.

Big AI performance. Tiny footprint.

HP ZGX Nano brings NVIDIA GB10 Grace Blackwell AI compute to the desk, lab, or edge environment in a compact desktop form factor. With 128GB of unified memory, 4TB of NVMe storage, and high-speed networking, it gives developers a practical local system for prototyping, fine-tuning, inference, and keeping sensitive AI workloads close to where the work happens.

Connectivity built for local AI environments

HP ZGX Nano combines high-speed networking, flexible display output, and USB-C expansion in a compact desktop AI station.

  • Dual 200G QSFP networking: Two NVIDIA ConnectX-7 200GbE interfaces support high-bandwidth connections to compatible systems or network infrastructure.
  • 10Gb Ethernet: A dedicated RJ45 port supports wired management and conventional high-speed networking.
  • USB-C expansion: Three rear USB-C ports support 20Gbps signaling and DisplayPort 1.4a Alt Mode for compatible displays and peripherals.
  • Display output: HDMI 2.1a provides a direct display connection, with audio output supported through HDMI.
  • Wireless connectivity: Wi-Fi 7 and Bluetooth 5.4 provide additional local connectivity where included with the regional configuration.

SEE IT IN ACTION

Meet the product

TECHNICAL DETAILS

Built into the system

NVIDIA GB10 Grace Blackwell AI compute

The NVIDIA GB10 Grace Blackwell Superchip combines a 20-core Arm CPU with NVIDIA Blackwell GPU architecture to provide compact, local AI compute. It delivers up to 1,000 TOPS of FP4 AI performance for compatible prototyping, fine-tuning, inference, and edge AI workloads.

128GB coherent unified memory

128GB of onboard LPDDR5X coherent unified memory is shared across the system architecture, helping support larger local datasets and compatible model workflows without relying on separate CPU and GPU memory pools. Memory bandwidth is rated at up to 273GB/s.

4TB fast, self-encrypting NVMe storage

This configuration includes a 4TB PCIe Gen4 NVMe OPAL M.2 SSD. The high-capacity local storage is suited to AI models, datasets, development environments, and project files while helping keep sensitive data on-premises.

Dual 200G networking with ConnectX-7

Two QSFP 200G interfaces powered by NVIDIA ConnectX-7 provide high-speed connectivity for compatible AI systems and network fabrics. The ZGX Nano can be paired with a second compatible ZGX Nano system to expand local compute resources for larger model workflows.

NVIDIA DGX OS and HP ZGX Toolkit

NVIDIA DGX OS provides a Linux-based AI software environment, while HP ZGX Toolkit brings together curated open-source tools, model discovery, tracking, export, and local serving. Together, they help reduce friction between experimentation and deployment.

Compact desktop design

The HP ZGX Nano measures approximately 5.9 × 5.9 × 2.1 inches and weighs about 2.76 lb. It uses a 240W external USB-C power adapter, making it a compact option for desks, labs, and edge environments.