NVIDIA offers HGX
Product
Modular reference platform letting server makers build NVLink-connected GPU systems for hyperscale AI and HPC.
- Currently Offered
Modular reference platform letting server makers build NVLink-connected GPU systems for hyperscale AI and HPC.
HGX is NVIDIA's data-center platform for artificial intelligence and high-performance computing, combining GPUs with high-speed interconnects, networking, and software.
Current systems include HGX Rubin NVL8, which pairs eight Rubin GPUs with sixth-generation NVLink. NVIDIA positions HGX for agentic AI, data analytics, and high-performance computing workloads.
The HP ZGX Nano AI Station is a compact desktop AI development system from HP.
Supermicro server systems powered by NVIDIA RTX PRO Blackwell Server Edition GPUs for enterprise AI, HPC, and visual computing.
HPE Cray XD670 is a 5U AI training server supporting eight NVIDIA H100 or H200 SXM5 GPUs.
Hyperstack is a European GPU-as-a-service cloud platform that gives developers and businesses API-first access to enterprise-grade infrastructure.
The QuantaGrid D75L-3U is a 3U GPU server carrying eight NVIDIA HGX B300 GPUs per node for AI workloads.
The ASUS Ascent GX10 is a compact desktop AI supercomputer built on the NVIDIA GB10 superchip.
Lenovo ThinkSystem SR675 V3 is a GPU-optimized rack server for enterprise artificial intelligence and professional graphics workloads.
PowerEdge XE9680 is Dell's 6U GPU-dense server for AI/ML supporting up to eight NVIDIA, AMD, or Intel accelerators.
Verified-counterparty venue trading GPU hardware, reserved rental, and forward contracts.
Dell Pro Max with GB10 is Dell's compact desktop AI development system built on the NVIDIA GB10 superchip.
Dedicated physical GPU servers without virtualization overhead for maximum AI performance.
Cerebras CS-4 is a rack-scale AI system built for fast, large-scale inference.
The HP ZGX Nano AI Station is a compact desktop AI development system from HP.
Supermicro server systems powered by NVIDIA RTX PRO Blackwell Server Edition GPUs for enterprise AI, HPC, and visual computing.
HPE Cray XD670 is a 5U AI training server supporting eight NVIDIA H100 or H200 SXM5 GPUs.
Hyperstack is a European GPU-as-a-service cloud platform that gives developers and businesses API-first access to enterprise-grade infrastructure.
The QuantaGrid D75L-3U is a 3U GPU server carrying eight NVIDIA HGX B300 GPUs per node for AI workloads.
The ASUS Ascent GX10 is a compact desktop AI supercomputer built on the NVIDIA GB10 superchip.
Lenovo ThinkSystem SR675 V3 is a GPU-optimized rack server for enterprise artificial intelligence and professional graphics workloads.
Cray XD670 is an HPE GPU-dense server for AI training and HPC.
PowerEdge XE9680 is Dell's 6U GPU-dense server for AI/ML supporting up to eight NVIDIA, AMD, or Intel accelerators.
Verified-counterparty venue trading GPU hardware, reserved rental, and forward contracts.
SambaNova DataScale is a data-center AI computing platform.
Dell Pro Max with GB10 is Dell's compact desktop AI development system built on the NVIDIA GB10 superchip.
Dedicated physical GPU servers without virtualization overhead for maximum AI performance.
Cerebras CS-4 is a rack-scale AI system built for fast, large-scale inference.
