GPU NODES

NVIDIA Accelerated GPU Nodes

Unlock high-performance AI, ML, and HPC workloads on bare-metal infrastructure accelerated by NVIDIA DGX Vera Rubin NVL72, GB300 NVL72, H100, H200, and GB200 GPUs — delivering unmatched scalability, energy efficiency, and fully integrated enterprise solutions

Performance

3,600 PFLOPS

NVFP4 inference performance

Built to accelerate large-scale AI inference workloads.

Get a quote
2,520 PFLOPS

NVFP4 training performance

Engineered for frontier-scale model training and post-training.

Learn more
75 TB

Total fast memory

Massive memory capacity for reasoning, simulation, and large AI workloads.

Talk to sales
72 Rubin GPUs

Unified by sixth-generation NVLink

Connected as one massive GPU fabric for rack-scale performance.

Learn more

No frills, just GPU compute

We built our GPU Nodes service for those who want the raw performance of bare metal GPUs. Say good bye to bloated infrastructure environments and hello to simplicity. Choose your node type and quantity, we’ll handle the rest.

Infrastructure that grows with you

All of our services are integrated with and built on the same infrastructure. This means you can add, remove, scale up, scale down your compute as your needs change. Start with bare metal GPU nodes and know you can layer on orchestration, scheduling, or application services in the future.

OUR GPUS

Built with industry leading accelerators

GB300 NVL72

Learn more

Engineered for massive-scale AI factories with next-generation Blackwell Ultra performance and unified NVLink architecture.

DGX Vera Rubin NVL72

Learn more

Purpose-built for gigascale AI training and inference with next-generation NVIDIA Vera Rubin architecture.

NVIDIA GB200 NVL72

Learn more

Designed for a new type of data center—the AI factory.

NVIDIA H200

Learn more

Supercharging AI and HPC workloads with NVIDIA H200 Tensor Core GPUs.

NVIDIA H100

Learn more

Experience the pinnacle of AI performance with NVIDIA H100 GPUs available.