AI Supercomputing Platform
Massive datasets, complex simulations, and advanced models require multiple GPUs linked by high-bandwidth interconnections to efficiently process and analyze data in real-time, ensuring high performance and scalability across simultaneous AI jobs at scale. The NVIDIA HGX™ AI supercomputing platform meets this demand by combining 8 NVIDIA GPUs with NVIDIA NVLink™ and NVIDIA networking into a unified architecture, delivering peak compute performance and faster time to insight for any data center.
Platform Overview
NVIDIA HGX Rubin NVL8
The NVIDIA HGX™ Rubin NVL8 scales AI factories for enterprise - through the co-design of Rubin GPUs, NVLink 6, ConnectX-9, BlueField-4, and Spectrum-X - to deliver fully optimized AI and high-performance computing (HPC) with accelerated time to insight.
Featured Product
The Ingrasys HGX™ Rubin NVL8 integrates advanced compute, HBM4 memory, and next-generation networking into an accelerated computing platform purpose-built for enterprise AI factories and high-performance computing (HPC) workloads. Featuring 100% direct liquid cooling across GPUs, CPUs, and key components, the system sustains peak performance and superior energy efficiency at the highest computational demands.
IGS-HXR200
- GPU: 8x NVIDIA Rubin GPUs
- CPU: 2x Intel® Xeon® 6 Processors
- Drives: 8x E1.S, 2x M.2 NVMe
- Expansion Slots: 1x PCIe 5.0 x 16 FHHL Slot
- Management: DC-SCM 2.0 (AST2600)
- Cooling Solution: Liquid Cooling Solution
- Dimensions (H x W x D): 3.42” x 17.63” x 31.4” / 87 x 448 x 800mm
_20260529__26F08xmEGO.png "HGX NVL8(BF4)_20260529")
