NVIDIA GB300 NVL72
Liquid Cooled Rack-Scale Solution with 72 NVIDIA B300 GPUs and 36 Grace CPUs
Checkout using your account
This form is protected by reCAPTCHA - the Google Privacy Policy and Terms of Service apply.
Checkout as a new customer
Creating an account has many benefits:
As enterprises accelerate AI adoption and high-performance computing (HPC) becomes mainstream, Supermicro’s GPU-optimised systems deliver unmatched flexibility, density and performance. Build your next-generation AI platform, fine-tune large language models (LLMs), enable high-fidelity 3D visualisation or deploy edge inference at scale — Supermicro GPU systems provide the hardware foundation.

Supermicro’s Building Block architecture allows flexible combinations of CPUs, GPUs, and memory—scaling from compact workstations to high-density AI clusters.
Optimised for demanding compute workloads with support for multi-GPU, NVLink, and high-bandwidth interconnects..
Compatible with the latest Intel® Xeon® and AMD EPYC™ processors, and a wide range of NVIDIA® and AMD Instinct™ GPUs.
Designed for large-scale environments with PCIe Gen5, high-speed networking, and efficient air or liquid cooling solutions.
Train large language models, generative AI systems or deep-learning workloads using high-density GPU nodes.
Deploy inference engines, multi-tenant AI services or real-time vision systems with GPU-accelerated infrastructure.
Run computational fluid dynamics (CFD), finite element analysis (FEA) or other HPC applications leveraging GPU-driven parallelism.
Support modern engineering, architecture, simulation and media workflows with multi-GPU workstations or rack servers.
Bring inference and intelligent workloads closer to the point of action with compact GPU servers suited for edge locations.
Liquid Cooled Rack-Scale Solution with 72 NVIDIA B300 GPUs and 36 Grace CPUs
Modular Building Block Design, Future Proof Open-Standards Based Platform for Large Scale AI training and HPC Applications
Direct-to-chip liquid-cooled systems for high-density AI infrastructure at scale.
Maximum Acceleration and Flexibility for AI, Deep Learning and HPC Applications
Modular Building Block Platform Supporting Today's and Future GPUs, CPUs, and DPUs
Multi-processor system combining CPU and GPU, Designed for the Convergence of AI and HPC
Flexible Design for AI and Graphically Intensive Workloads, Supporting Up to 10 GPUs
Dense and Resource-saving Multi-GPU Architecture for Cloud-Scale Data Center Applications
Let’s talk Supermicro. Fill out the form below to get expert guidance, custom quotes, and deployment-ready solutions tailored to your business.