Full rack with 72 NVIDIA Blackwell GPUs, 36 NVIDIA Grace CPUs and 5th-gen NVIDIA NVLink, enabling trillion-parameter LLM inference and training. Supports liquid cooling and scale-up ecosystem-ready.

Key Features

• Unparalleled AI performance: A powerhouse that contains 72 NVIDIA Blackwell GPUs and 36 Grace CPUs for exceptional AI inference and training.
• Superior efficiency: The full-rack solution is a symphony of GPUs, CPUs and switches harmonizing in lightning-fast, direct communication, turbocharging trillion-parameter LLM training and real-time inference.
• High-speed interconnect: Features fifth-generation NVIDIA NVLink technology, 14.4TB/s switching capacity for high-bandwidth, low-latency GPU connections within a single rack, enabling seamless communication and superior throughput.
• Advanced cooling portfolio: Supports both liquid-to-air and liquid-to-liquid cooling solutions, unleashing optimal AI computing performance while ensuring effective heat dissipation.