Enterprise Infrastructure for Mission-Critical AI

NVIDIA DGX SuperPOD™ with DGX GB200 systems is purpose-built for training and inferencing trillion-parameter generative AI models. Each liquid-cooled rack features 36 NVIDIA GB200 Grace Blackwell Superchips–36 NVIDIA Grace CPUs and 72 Blackwell GPUs–connected as one with NVIDIA NVLink. Multiple racks connect with NVIDIA Quantum InfiniBand to scale up to tens of thousands of GB200 Superchips.

Benefits

Enterprise Generative AI Infrastructure
With Constant Uptime

Maximize Developer Productivity

An intelligent control plane tracks thousands of data points across hardware, software, and data center infrastructure to ensure continuous operation and data integrity, plan for maintenance and automatically reconfigures the cluster to avoid downtime.

Massive Supercomputing for Generative AI

Scaling up to tens of thousands of NVIDIA GB200 Superchips, DGX SuperPOD with DGX GB200 systems effortlessly performs training and inference on state-of-the-art trillion-parameter generative AI models.

Built on NVIDIA Grace Blackwell

NVIDIA GB200 Superchips, each with one Grace CPU and two Blackwell GPUs, are connected via fifth-generation NVLink to achieve 1.8 terabytes per second (TB/s) of GPU-to-GPU bandwidth.