The A rack in the center, with four server products positioned beside it, and a background featuring AI, semiconductor chip, and DNA-inspired visuals. A rack in the center, with four server products positioned beside it, and a background featuring AI, semiconductor chip, and DNA-inspired visuals.

ASUS NVIDIA-Certified AI Servers, Storage & SoftwareASUS NVIDIA-Certified AI Servers, Storage & Software

ASUS brings accelerated computing to scale with NVIDIA-Certified Systems™

ASUS and NVIDIA® are at the forefront of AI and accelerated computing, delivering powerful solutions that empower businesses to innovate and excel. The offerings include advanced hardware integration, with layouts for switches, storage, racks, cooling systems, software, as well as server clusters and cloud platforms, supporting a wide array of applications and ensuring optimal performance and flexibility.

Imagine leveraging the latest technology to unlock new possibilities in data science and machine learning. ASUS and NVIDIA's software platforms are meticulously designed to streamline workflows, making it easier than ever to develop, deploy, and manage AI projects. Whether you're building intelligent applications or analyzing vast datasets, these solutions provide the tools you need to stay ahead in today's competitive landscape.

Explore how these pioneering technologies can transform your approach to AI and elevate your organization to new heights

The NVIDIA logo is positioned above, with Certified enclosed in a framed design.

Why consider an NVIDIA-Certified Systems™ ?

ASUS servers have been tested and verified by NVIDIA. With the certification from NVIDIA, it means the servers are designed to meet rigorous performance and reliability standards, particularly for resource-intensive applications such as AI, deep learning and data analytics. These systems are built with cutting-edge NVIDIA GPUs and networking hardware, ensuring top-tier performance in standardized software environments.

ASUS qualified systems 〉

Your best choice of computing from edge to data center

Scale Out
A front-view photo of an open-chassis server

NVIDIA L40/L40S GPU, NVIDIA H200 NVL, NVIDIA RTX PRO™ 6000 Blackwell Server Edition, NVIDIA RTX PRO™ 4500 Blackwell

Scale Up
front and rear photos of the ASUS AI POD with NVIDIA VERA RUBIN NVL72 rack.

NVIDIA HGX™ B300, NVIDIA Grace Blackwell, NVIDIA GB200 NVL72, NVIDIA GB300 NVL72, NVIDIA Vera Rubin NVL72

405B~1T
175B
70B
7B
Parameter model size
NVIDIA MGX™ 〉 NVIDIA OVX™ 〉
NVIDIA NVL72 〉 NVIDIA HGX™ 〉 NVIDIA CMX 〉
NVIDIA Solutions 〉

ASUS AI POD with NVIDIA NVL72 Platform

ASUS AI POD solutions power next-gen AI factories and trillion-parameter models using the NVIDIA Vera Rubin and NVIDIA Grace Blackwell platform. These meticulously engineered systems feature advanced hardware (processors, networks, storage) and come with a complete turnkey service, covering deployment from data center setup to onboarding, ensuring peak performance and reliability.

XA VR721-E3
NVIDIA Vera Rubin NVL72

Building the Next Frontier of AI

  • 36 x NVIDIA Vera CPUs
  • 72 x NVIDIA Rubin GPUs
  • NVIDIA Technology:6th Gen NVIDIA NVLink™,
  • NVIDIA ConnectX®-9 SuperNIC™, NVIDIA BlueField®-4 DPU
  • Cable-less and 100% liquid-cooled design
  • Modular and SOCAMM design for serviceability
  • ASUS Infrastructure Deployment Center
  • ASUS Professional Services
Learn More 〉
XA VR721-E3 NVIDIA Vera Rubin NVL72

ASUS NVIDIA HGX™ systems

With its integration of NVIDIA GPUs, NVIDIA HGX platforms deliver unparalleled high-performance computing capabilities and provides the robust infrastructure you need that redefine what's possible in the world of AI. Whether you're developing complex AI models or powering real-time applications, the HGX platforms provide the robust infrastructure you need to unleash creativity and efficiency.

XA NR1I-E12L
NVIDIA HGX™ Rubin NVL8

Large-scale AI Model Training & Inference

  • Powered by dual Intel® Xeon® 6700P processors, supporting a 350W TDP
  • NVIDIA HGX™ Rubin NVL8, providing 800Gb/s bandwidth for E/W bandwidth per GPU (OSFP)
  • Seamless GPU-to-GPU interconnect via NVIDIA® NVLink™, providing 3600GB/s bandwidth for optimized scalability
  • Independent CPU and GPU sled for easy maintenance
  • Adopting an innovative hybrid cooling design that combines D2C liquid cooling with a 100% air-cooled CPU sled can reduce maintenance time by 50% during servicing.
Learn More 〉
XA NR1I-E12L NVIDIA HGX™ Rubin NVL8

ASUS NVIDIA MGX™ systems

Engineered with an innovative modular architecture, ASUS NVIDIA RTX Pro servers offer unparalleled flexibility and scalability to accommodate various server configurations - empowering clients to tailor AI solutions to precise needs. MGX™ provides a new standard for modular server design by improving return on investment (ROI) and reducing time to market.

ESC8000A-E13P
NVIDIA H200 NVL PCIe, NVIDIA RTX PRO™ 6000 Blackwell Server Edition, NVIDIA RTX PRO™ 4500 Blackwell Server Edition

Turbocharging generative AI and LLM workloads

  • Powered by AMD EPYC™ 9005 processors with 192 Zen 5c cores, 12-channel memory, DDR5 up to 6400MHz, and a maximum TDP of 500W per socket
  • Fully compatible with NVIDIA MGX™ architecture, enabling rapid and scalable deployment
  • High-density 4U server for eight dual-slot high-end NVIDIA H200 GPUs or NVIDIA RTX PRO™ 6000 Blackwell Server Edition, NVIDIA RTX PRO™ 4500 Blackwell Server Edition, supporting up to 600 watts each
  • Optimized configuration with five PCIe 5.0 slots for high-bandwidth NICs and DPUs, ensuring scalable performance
  • ASUS-exclusive toolless design for simplified maintenance and operational efficiency
  • ASUS Control Center Data Center Edition, paired with a hardware-level Root-of-Trust solution, provides enhanced security.
Learn More 〉 Data Sheet 〉
ESC8000A-E13P NVIDIA H200 NVL PCIe, NVIDIA RTX PRO™ 6000 Blackwell Server Edition, NVIDIA RTX PRO™ 4500 Blackwell Server Edition

ASUS NVIDIA OVX™ systems

NVIDIA OVX™ is a purpose-built computing platform designed to power large-scale, physically accurate, and collaborative 3D simulation and digital twin workloads. It combines high-performance GPUs, high-speed networking, and optimized software to accelerate the NVIDIA Omniverse™ platform, enabling creators, engineers, and developers to build and operate industrial-scale digital worlds. NVIDIA OVX™ is ideal for industries such as architecture, engineering, manufacturing, and entertainment, where real-time collaboration and simulation precision are critical for design and decision-making processes.

ESC 8000A-E13P
NVIDIA L40S PCIe

Turbocharging generative AI and LLM workloads

  • Powered by AMD EPYC 9005 processors with up to 192 Zen 5c cores, 12-channel DDR5 at up to 6400 MHz, and a maximum TDP of 500W per socket
  • High-density 4U server supporting NVIDIA L40S PCIe with up to eight dual-slot 600W GPUs, NVIDIA® NVLink™ 2-Way or 4-Way bridges, and BlueField®-3 DPUs for scalable performance
  • Optimized with five PCIe 5.0 slots for high-bandwidth NICs and DPUs, enhancing network throughput and system performance
  • Enhanced air cooling based on CPU/GPU TDP for diverse workloads
  • ASUS-exclusive toolless design empowers easy maintenance for maximum efficiency
  • ASUS Control Center Data Center Edition, paired with a hardware-level Root-of-Trust solution, provides enhanced security.
Learn More 〉 Data Sheet 〉
ESC 8000A-E13P NVIDIA L40S PCIe

ASUS NVIDIA CMX Storage Server

ASUS context memory (CMX) storage server integrates NVIDIA's modular STX architecture, providing a unified hardware and software infrastructure. Optimized for complex and generative AI workloads, this specialized design transcends conventional servers to meet next-generation AI system demands.

UF920-E3-RS24
Validated storage server
  • Empowering by NVIDIA Vera CPU with NVIDIA BlueField-4 DPU and ConnectX-9 SuperNIC technology.
  • Optimized end-to-end AI data pipelines by connecting ASUS CMX with NVIDIA GPUs, reducing bottlenecks and maximizing GPU utilization.
  • Strategic Software Partnerships with WEKA and IBM to integrate advanced data management and intelligent services
Learn More 〉
UF920-E3-RS24

NVIDIA Solutions

All ASUS GPU servers carry NVIDIA-Certified status so they're ready for NVIDIA solutions, ensuring complete compatibility and interoperability with the entire ecosystem.

NVIDIA DSX

NVIDIA DSX is an AI factory-scale platform, unifying design, simulation, operations, and ecosystem technologies. It defines how AI factories are designed, built, and optimized across the full stack, from chips and systems to infrastructure software, facilities, and partner technologies. DSX combines open software libraries, workflow guides, and reference designs with NVIDIA compute platforms and codesigned infrastructure to enable a broad ecosystem of software and hardware solutions. By aligning every layer through a common architecture, DSX improves tokens per watt, accelerates deployment, and strengthens operational reliability and resiliency.

Learn More 〉
NVIDIA AI Enterprise

ASUS and NVIDIA® deliver the industry's first full-stack to help you tune AI models more securely, ready for seamless deployment of AI applications. Our diverse range of configuration options can support NVIDIA AI Enterprise (AIE). AIE is a cloud-native software platform that accelerates the development and deployment of AI applications. It streamlines data science pipelines and enhances productivity in AI-driven environments.

Learn More 〉
NVIDIA Omniverse™ Enterprise

ASUS GPU servers combined with NVIDIA Omniverse™ Enterprise offer an unmatched solution for industries looking to enhance all kinds of workflows. With real-time collaboration, enhanced processing power, and integrated tools, you can unlock new levels of productivity and innovation. NVIDIA Omniverse Enterprise is realizing digital-twin workflows by providing an open platform that fosters real-time, collaborative design across multiple industries. This platform connects global teams, allowing them to work together seamlessly, improving creativity and efficiency. If we take manufacturing automation as an example, Omniverse can deploy sensors and devices across the physical twin, we can capture a continuous stream of real-time data, such as temperature, vibration, and operational performance.

Learn More 〉
NVIDIA Run:ai

ASUS GPU servers with NVIDIA Run:ai optimize AI infrastructure by automating compute resource management. This powerful orchestration layer provides advanced scheduling and fractional GPU sharing, enabling efficient utilization across single GPUs or large clusters. It boosts operational efficiency, maximizes ROI by eliminating idle resources, and accelerates AI initiatives from development to production within multi-tenant environments.

Learn More 〉

ASUS Professional Services

AI servers from ASUS serve as ideal platforms for AI services, ensuring seamless compatibility and operation for any AI applications clients wish to deploy. ASUS provides assistance in service scheduling, resource monitoring, and ensuring smooth operations on top of ASUS hardware. The expertise of specialist ASUS teams minimizes errors and downtime by preventing costly misconfigurations, ensuring that your servers run smoothly and reliably. With an intuitive interface and powerful automation features, ASUS simplifies server management, making it effortless to handle complex tasks.

  • In a circle, there are two people, one sitting and one standing.

    Consult & Design

    • Demand evaluation
    • Configuration and architecture design
    • Computing and network topology
    • Cooling, Electricity, layout design
  • In a circle, three gears are interlocked.

    Install

    • Server/switch set up
    • Cabling and labeling
    • Electrical/UPS
    • In-row CDU/ Sidecar
    • Fire/Safety system
  • In a circle, there is a document and a pen, with a checkmark displayed on the document.

    Validation

    • Acceptance test and accept reporting
    • System foundation validation
    • Port mapping
    • Robust interconnectivity
  • In a circle, there is a document containing a flowchart.

    Deployment (AIDC)

    • Rapid deployment
    • Centralized configuration
    • Automation & systemization
    • GenAI cloud services
  • In a circle, there is a book with DIY tool illustrations on it.

    Maintain & Support

Learn More 〉
Two rows, each with eight NVIDIA GB200 NVL72 units, paired with 2 CDUs. Cold water pipelines are interwoven throughout.

ASUS, your trusted AI infrastructure partner

Data center operations are vital to modern business infrastructure. After acquiring servers, it's crucial to consistently monitor and manage factors like efficiency, stability and scalability to meet growing demands. Effective data center management involves more than just selecting and configuring hardware: it requires holistic oversight of networking, security, maintenance, scalability and cost control. A well-rounded approach ensures optimal performance and long-term stability.

When you choose ASUS, you're not merely acquiring hardware, you're optimizing the total cost of ownership (TCO) and improving operational efficiency. Our team of performance tuning experts works behind the scenes to ensure your systems are running at their peak performance. This maximizes your return on investment (ROI) and enhances long-term success. ASUS provides a complete solution that goes beyond just hardware, addressing all aspects of your digital infrastructure needs.

  • Two lines winding around the Earth.
    Global Services and Strong Ecosystem

    2,500+ global service coverage and tailor-made SLA/end to end solution services & work closely with major vendor and partners worldwide

  • Three servers stacked with a shield-shaped symbol in front.
    L10-L12 Capability

    A leading provider of comprehensive hardware/software solutions, offering flexible assembly support across key regions including the US, Taiwan and more.

  • A cloud with three lines below it, connecting different blocks.
    Cloud Readiness

    Datacenter cloud computing operation & management capabili

  • A gear with three lines below it.
    Customization

    In-house EE/ME/SW/thermal/power/rack design/logistic solution

FAQs

ASUS NVIDIA-Certified Systems are designed for AI training, generative AI, deep learning, high-performance computing (HPC), data analytics, and enterprise virtualization workloads. These systems combine advanced NVIDIA GPUs, networking, and ASUS server infrastructure to deliver scalable AI performance from edge computing to modern data centers.

ASUS NVIDIA-Certified servers are tested and validated to meet strict NVIDIA standards for performance, scalability, reliability, and security. Businesses benefit from optimized AI infrastructure, faster deployment, improved GPU acceleration, enterprise-grade thermal design, and support for demanding workloads such as large language models (LLMs), AI inference, and Omniverse applications.

ASUS supports a wide range of NVIDIA AI platforms including NVIDIA Vera Rubin, NVIDIA Blackwell, NVIDIA GB200 NVL72, NVIDIA GB300 NVL72, NVIDIA HGX™ Rubin, NVIDIA HGX™ B300, NVIDIA H200 NVL, NVIDIA L40S, and NVIDIA RTX PRO™ Blackwell Server Edition GPUs. These platforms support scalable AI deployments for cloud providers, enterprises, and research organizations.

ASUS AI POD solutions are engineered for large-scale AI model training and inference using high-density GPU clusters, advanced networking, optimized cooling, and turnkey deployment services. They help organizations accelerate agentic AI, generative AI, AI reasoning, digital twins, and LLM workloads while simplifying deployment and infrastructure management.

Yes. ASUS offers end-to-end AI infrastructure services including consultation, architecture design, installation, validation, deployment, monitoring, maintenance, and lifecycle support. ASUS Professional Services help businesses optimize data center efficiency, reduce downtime, and accelerate AI adoption with customized enterprise solutions.