Hit the ground prototyping, fine-tuning, and inferencing with familiarNVIDIA® DGXTM OS and integrated NVIDIA® AI software stack purpose-built for modern AI development.
NVIDIA® GB10 GraceBlackwell Superchip
Get up to 1000 TOPS of AI compute at FP4 precision with an NVIDIA®Blackwell GPU. Supercharge data preprocessing and orchestration with aGrace 20-core Arm CPU.[3]
Unified systemmemory
Run AI development and testing workloads with AI models of up to 200billion parameters at your desk with 128 GB of coherent unified systemmemory.
NVIDIA® ConnectX™ Networking
Work with even larger AI models locally—up to 405 billion parameters—by connecting two HP ZGX Nano systems together to scale localcompute resources.[4]
Tiny AI powerhouse
Measuring at 150mm L x 150mm W x 51mm H,5 this new class ofdesktop is purpose-built for AI development yet fits in the palm of yourhand.[5]
Big AI performance. Tiny footprint.
Prototype, fine-tune, and inference models of up to 200B parameters locally. Powered by NVIDIA® GB10 Grace Blackwell Superchip[2] and 128 GB of coherent unified memory, the HP ZGX Nano delivers 1,000 TOPS of FP4 AI performance in a compact desktop.
HP ZGX Toolkit: Accelerate time to results
Move from idea to deployment faster. The ZGX Toolkit[1] streamlines AI workflows by pairing powerful local compute with open-source tools, built-in discovery, and easy export—helping teams cut friction, boost productivity, and scale results anywhere.
Offload AI workloads without the cloud
Pair existing laptop and desktop systems—Windows, Mac, or Linux—with a network-connected HP ZGX Nano. Get developer-grade performance without data center queues and costly cloud instances, minimizing latency and keeping sensitive information local.
Advance edge AI solutions
Realize the true potential of edge AI solutions. From real-time computer vision to running agentic and domain-specific models locally, get the performance needed to deploy intelligent, responsive solutions directly where data is generated.
AI supercomputing goes Nano Accelerate AI workflows with HP ZGX Nano and ZGX Toolkit[1], high-performance local compute combined with a curated open-source stack for prototyping, fine-tuning and inferencing. Built-in IP discovery, model export and local serving reduce friction, ensuring repeatable, deployment-ready results.
