Overview
The Foundation for Intelligence at Scale
Agentic AI and reasoning models are redefining the limits of computation. But to take full advantage of these breakthroughs, enterprises need infrastructure that can reason internally and quickly pass information between models and agents. Infrastructure must cost-effectively address communication, coordination, memory movement, and latency across training, post-training, and inference.
NVIDIA DGX™ Rubin NVL8 provides a blueprint for success in the age of agentic AI. Built on the NVIDIA Rubin architecture, DGX Rubin NVL8 is a turnkey infrastructure solution purpose-built to accelerate any AI workload and deliver intelligence at scale.
Features
Scaling Intelligence for the Age of Reasoning
Supercharged Performance for Agentic AI
Powered by eight NVIDIA Rubin GPUs, DGX Rubin NVL8 delivers 400 petaFLOPS of NVFP4 performance and 176 terabytes per second (TB/s) of GPU memory to transport data as quickly as possible. DGX Rubin NVL8 leverages sixth-generation NVIDIA NVLink™ to ensure seamless peer-to-peer communication for massive model parallelism.
Powered by NVIDIA Rubin Architecture
DGX Rubin NVL8 dramatically improves economic viability by optimizing performance across the entire AI lifecycle. It features a specialized multi-agent engine for reasoning workflows and a dedicated reinforcement learning engine that optimizes memory movement in hardware.
Accelerating Operations With Integrated Software
NVIDIA Mission Control™ accelerates every aspect of operations, from configuring DGX Rubin NVL8 to integrating with facilities to managing clusters and workloads. Enabled by intelligent, integrated software, enterprises gain control over cooling and power events to redefine infrastructure resiliency.
The Proven Standard for Enterprise AI
Built from the ground up for enterprise AI, the NVIDIA DGX platform, featuring NVIDIA DGX SuperPOD, combines the best of NVIDIA software, infrastructure, and expertise in a modern, unified AI development solution.
Specifications
NVIDIA DGX Rubin NVL8 Specifications
| GPU | 8x NVIDIA Rubin GPUs |
| Total GPU Memory | Bandwidth | 2 TB | 130 TB/s |
| Performance | NVFP4 Inference: 400 PFLOPS*
NVFP4 Training: 270 PFLOPS** FP8/FP6 Training: 130 PFLOPS** |
| CPU | 2x Intel® Xeon® 6776P processors |
| NVIDIA NVLink Switch Bandwidth | 24 TB/s total bandwidth |
| System Power Usage | ~24 kW |
| Networking | 8x OSFP ports serving 8x single-port NVIDIA® ConnectX®-9 VPI
|
| Software | NVIDIA DGX OS, Ubuntu, Red Hat Enterprise Linux, Rocky |
|
*Sparse specification.
|
|
Offers
Delivering Supercomputing to Every Enterprise
Get Started
Learn More About NVIDIA DGX Rubin NVL8
Speak to an NVIDIA expert about NVIDIA DGX Rubin NVL8.
Discover the Benefits of the NVIDIA DGX Platform
Learn why NVIDIA DGX is the proven standard on which enterprise AI is built.
Explore the NVIDIA DGX Ecosystem
Unlock the full potential of AI with flexible deployment options, from on-premises to hosted solutions.