Hardware

AI infrastructure hardware — data-center GPUs, accelerators, and the networking that ties clusters together. Browse the catalog and request a quote for current availability and configuration.

• NVIDIA

  NVIDIA H100 SXM

  NVIDIA Hopper-architecture data-center GPU in the SXM5 module form factor, built for large-scale AI training and inference.

• NVIDIA

  NVIDIA H200

  NVIDIA Hopper-architecture data-center GPU with HBM3e memory, positioned for memory-bound large-language-model inference and training.

• NVIDIA

  NVIDIA A100 80GB

  NVIDIA Ampere-architecture data-center GPU with 80GB of HBM2e memory, a widely deployed accelerator for AI training, inference, and HPC.

• AMD

  AMD Instinct MI300X

  AMD CDNA-architecture data-center accelerator with high-capacity HBM3 memory, built for large-language-model training and inference.

• NVIDIA

  NVIDIA Quantum-2 InfiniBand

  NVIDIA Quantum-2 InfiniBand switching platform for the high-bandwidth, low-latency fabric that interconnects GPU clusters at scale.