XA GB721-E2

ASUS AI POD powered by NVIDIA GB300 NVL72

Engineered for the most demanding AI workloads, the ASUS AI POD, powered by the NVIDIA® GB300 NVL72, delivers unparalleled compute density and efficiency. This robust platform is purpose-built to accelerate large-scale high performance computing, sophisticated reasoning, and drive transformative breakthroughs in AI.
A rack is configured with multiple compute trays. The ASUS AI with NVIDIA GB300 NVL72
See What Powers It right arrow Get Personalized Pricing right arrow
  • A dark blue card with a gear and rising bar-chart icon

    Predictable Performance​

    Assured bandwidth and low latency for demanding AI workloads​

  • A dark blue card with a network-node icon

    Simplified Scaling​

    Validated designs ensure smooth growth from rack to cluster​

  • A dark blue card with a gear and connected-nodes icon

    Deployment Efficiency​

    Reference architectures accelerate setup and reduce complexity​

A tall open-panel ASUS AI POD ESC NM2N721-E1 NVIDIA GB200 NVL72

Built for the Era of AI Reasoning

ASUS AI POD with NVIDIA GB300 NVL72 combines 72 Blackwell Ultra GPUs and 36 Grace™ CPUs in a liquid-cooled, rack-scale platform designed to scale AI factories to unprecedented levels.
Learn More Details right arrow Download Datasheet right arrow
    • The front-view of ASUS AI POD with NVIDIA GB300 NVL72

      XA GB721-E2

      CPU
      36 x NVIDIA Grace CPUs
      GPU
      72 x NVIDIA Blackwell Ultra GPUs
      Networking
      2 × OOB MGMT Switch 1GBe
      9 × NVLink Switch Trays
      Cooling
      Support Liquid-to-liquid or Liquid-to-air
      Rack Power Shelf
      4 x 1RU 33kW power shelves

Your Trusted Partner for AI Infrastructure

As one of the first vendors in the world to deliver XA GB721-E2 with NVIDIA GB300 NVL72, ASUS leads in AI server infrastructure, offering reliable, efficient direct liquid cooling for data centers. Whether you're upgrading existing facilities or building new AI factories, ASUS provides proven liquid-cooled AI servers that accelerate deployment and maximize performance.
Two rows of eight ASUS AI POD with NVIDIA GB300 NVL72. Each row paired with one CDU, while cold-water pipelines are seamlessly integrated throughout the deployment.

Take a Closer Look

A wide-angle view of a modern data center showcasing rows of ASUS AI POD system

Product Video

NVIDIA GB300 NVL72 | ASUS AI POD
Watch now right arrow
ASUS AI POD cluster

Successful story

Accelerating AI Cloud Infrastructure with ASUS AI POD with NVIDIA GB300 NVL72, XA GB721-E2
Read more right arrow
Multiple chilled-water pipelines running throughout the data center.

Technical Insight

Inside the ASUS Infrastructure Lab- NVIDIA GB300 NVL72 Liquid Cooling
Read more right arrow
A vibrant ASUS exhibition booth, teeming with attendees exploring and interacting with the showcased technology.

Newsroom

ASUS Unveils the Latest ASUS AI POD Featuring NVIDIA GB300 NVL72 at GTC with Major orders secured
Read more right arrow

FAQ

  • What are the key benefits and infrastructure requirements of GB300 NVL72?
    Q1. What are the key benefits and infrastructure requirements of GB300 NVL72?NVIDIA GB300 NVL72 delivers significant advantages for large-scale AI workloads, including faster model training and inference, improved efficiency for generative AI applications, better scalability, and optimized total cost of ownership at scale.To fully realize its performance, it requires advanced AI infrastructure, including high-density rack design, high-bandwidth low-latency networking, high-capacity power delivery, and advanced liquid or hybrid cooling systems to ensure stable, sustained operation under intensive workloads.

Still have further questions?

Explore more right arrow