ASUS ExpertCenter Pro ET900N G3

Your Trusted AI Supercomputer

Now available

Overview

ASUS ExpertCenter Pro
ET900N G3

Based on NVIDIA DGX™ Station Architecture

The ASUS ExpertCenter Pro ET900N G3 is part of a new class of deskside AI supercomputers. Designed from the ground up to build and run AI, this revolutionary system unleashes the NVIDIA® GB300 Grace Blackwell Ultra Desktop Superchip and a massive 748GB of large coherent memory. Fully compatible with NVIDIA NemoClaw and NVIDIA AI Enterprise, it enables the local deployment of trillion-parameter models and autonomous AI agents at the deskside. Capable of developing and running large local AI training and inferencing workloads, this desktop system uses the NVIDIA® AI Software Stack to put an unprecedented amount of compute performance at the fingertips of AI development teams.

For more information see:

NVIDIA^® GB300 Grace Blackwell Ultra Desktop Superchip
748 GB

of Coherent Memory
Up to

20 PFLOPS

AI Performance
NVIDIA ConnectX^®-8 SuperNIC

Significantly enhances system performance for AI factories.
NVIDIA
NVLink™-C2C

Enables high-bandwidth, coherent data transfer between processors and accelerators.
Support

NVIDIA^® NemoClaw

For secure, streamlined AI agent workflows.
NVIDIA AI Software Stack

Empowers developers to create and execute AI workflows.
NVIDIA Multi-Instance GPU (MIG)

MIG partitions GPUs into up to seven isolated instances, enabling secure, efficient, and flexible usage.
* Preliminary specifications, subject to change.

Applications

Powering AI Across Industries and Beyond

Agentic AI

Secure and Scalable Agentic AI with NVIDIA NemoClaw

The ASUS ExpertCenter Pro ET900N G3 with the NVIDIA NemoClaw open source stack, adds privacy and security controls to OpenClaw. NemoClaw installs NVIDIA OpenShell to enforce policy-based privacy and security guardrails, giving users control over how agents behave and handle data and provides the compute resources to run open models like NVIDIA Nemotron locally for enhanced privacy and cost efficiency.

Thermal

Extreme Cooling for Extreme AI Workloads

Features a data center-grade thermal architecture designed to sustain extreme AI workloads without compromise. With optimized airflow, advanced heat dissipation, and precision thermal control, it efficiently supports next-generation deskside supercomputer performance for AI training and inference. This enables long-duration, full-load local AI computing with zero thermal throttling, consistent peak performance, and 24/7 reliability for mission-critical workloads.

Specifications

ASUS ExpertCenter Pro ET900N G3

Operating System	Ubuntu with NVIDIA AI Developer Tools
CPU	Grace 72-Core Neoverse V2
Graphics	NVIDIA Blackwell Ultra
Memory	748GB of Coherent Memory
Storage	Up to 4 x M.2 2280 NVMe PCIe 5.0 SSD, total 8TB
Wireless Data Network	Optional (WIFI)
LAN	1 x 10G LAN, 1 x 1G LAN (Dedicated management LAN for BMC)
Front(Side) I/O Ports	2 x USB 10Gbs Type-C ports 2 x USB 10Gbs ports 1 x USB 2.0 port 1 x Head phone port 1 x Microphone port
Back I/O Ports	4 x USB 10Gbps port (4 x Type-A) 1 x USB Micro B (for BMC) 1 x Marvel 10Gb Ethernet port 1 x Realtek 1Gb Ethernet port (for BMCC) 2 x CX-8 NIC ports 1 x mini DisplayPort (for BMC) 3x Audio jacks
Power Supply	1600W TITANIUM ATX Power Supply
Dimensions (W x D x H)	232 x 584 x 565 mm ( 9.1 x 23 x 22.3 inch)
Net Weight	27 kg
Gross Weight	32 kg
Certificates	CE, FCC (Class A), BSMI, UL, KCC

Frequently Asked Questions

What is a deskside AI supercomputer?

A deskside AI supercomputer delivers data center–class AI performance in a compact, office-ready system.
It enables local training and inference of large AI models with lower latency and greater data control.
The ASUS ExpertCenter Pro ET900N G3, powered by the NVIDIA® GB300 Grace Blackwell Ultra Desktop Superchip, brings this capability to the deskside for enterprise AI deployment.

What is the difference between cloud AI and on-prem AI?

Cloud AI runs on remote data centers, offering scalability but with potential latency, cost, and data privacy concerns.
On-prem AI runs locally, delivering faster performance, predictable costs, and full data control.
Systems like the ASUS ExpertCenter Pro ET900N G3 bring data center–class AI performance to the deskside, enabling large AI models to run locally.

What AI performance does ASUS ExpertCenter Pro ET900N G3 deliver?

The ET900N G3 delivers up to 20 PFLOPS of AI performanceK, powered by the NVIDIA® GB300 Grace Blackwell Ultra Desktop Superchip, enabling large-scale AI training and real-time inference at the deskside.

How much memory does ET900N G3 support?

The system features 748GB of large coherent unified memory, allowing developers to run trillion-parameter AI models and memory-intensive workloads locally.

Can ET900N G3 run large AI models locally?

Yes. With its unified memory architecture and NVIDIA AI software stack, ET900N G3 enables local deployment of large language models (LLMs) and AI agents without relying on cloud infrastructure.

What makes ET900N G3 different from traditional workstations?

Unlike traditional workstations, ET900N G3 is a deskside AI supercomputer designed specifically for AI workloads, combining data center-class compute, memory, and software in a desktop form factor.

What AI applications does ET900N G3 support?

ET900N G3 supports a wide range of AI workloads, including:
• Large language model (LLM) training & inference
• Autonomous AI agents
• Computer vision
• Generative AI content creation
• Scientific computing

What software ecosystem is supported?

The ASUS ExpertCenter Pro ET900N G3 is designed to support leading AI software ecosystems, including:
NVIDIA AI Enterprise
NVIDIA OpenShell
NVIDIA AI software stack
These platforms can be deployed on ET900N G3 to enable optimized performance across AI training, inference, and deployment workflows.

Is ET900N G3 suitable for continuous AI workloads?

Yes. With its data center-grade thermal architecture, ET900N G3 is designed for 24/7 operation, ensuring stable performance under sustained AI workloads.