Bringing data-center-class AI computing to enterprises, developers, and AI innovators at the deskside
- NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip: Delivers multi-petaflop-scale AI computing for advanced AI development and inference
- Massive coherent memory: 748GB coherent CPU-GPU memory enables larger AI model training and accelerated local AI workflows, capable of running frontier AI models up to 1 trillion parameters.
- Up to 20 PFLOPS AI performance: Data-center-class performance in a local system to accelerates LLM fine-tuning, generative AI, autonomous AI agents, simulations, and deep learning workloads
- Enterprise-ready AI infrastructure: Built on NVIDIA DGX Station architecture with enterprise-grade scalability, connectivity, and reliability
TAIPEI, June 23, 2026 /PRNewswire/ — ASUS today announced the availability of ASUS ExpertCenter Pro ET900N G3, a next-generation deskside AI supercomputer engineered to bring data-center-class AI performance directly to enterprises, AI developers, researchers, and data scientists. Powered by NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip and built on NVIDIA DGX Station GB300 architecture, ExpertCenter Pro ET900N G3 delivers exceptional AI compute, massive coherent memory capacity, and enterprise-grade AI infrastructure in a deskside form factor, with future support planned for Windows-based AI development and agentic environments.
As AI adoption rapidly expands across industries, organizations increasingly require powerful local AI infrastructure capable of supporting large-scale AI training, inference, and autonomous agentic AI workflows without relying entirely on cloud computing. ExpertCenter Pro ET900N G3 addresses this growing demand by enabling enterprises to deploy advanced AI capabilities locally with greater control, lower latency, predictable operational costs, and enhanced data privacy. The system is now available worldwide. To learn more or explore deployment opportunities, visit the Contact Me page for further information and consultation.
Built for next-generation AI development
ExpertCenter Pro ET900N G3 is designed to support modern AI workloads ranging from large language model (LLM) fine-tuning and generative AI to physical AI, deep learning research, and autonomous AI agents. Powered by NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip connected through NVIDIA NVLink-C2C high-bandwidth interconnect technology, the system enables high-speed coherent memory access for accelerated AI processing.
With 748GB of coherent unified memory, ET900N G3 enables developers and researchers to work with significantly larger AI models locally while reducing workflow bottlenecks traditionally associated with conventional workstations. Delivering up to 20 PFLOPS of AI performance, the system provides the computational power required for demanding AI development and inference workloads directly at the desktop.
Data-center-class AI infrastructure in a deskside form factor
Built on the NVIDIA DGX Station GB300 architecture, ASUS ExpertCenter Pro ET900N G3 brings supercomputer-level AI performance into an office-friendly deskside design. The system combines powerful AI acceleration with enterprise-class scalability and connectivity, enabling businesses and research institutions to rapidly deploy AI infrastructure without requiring dedicated data-center environments.
ET900N G3 is optimized for local AI development and supports the NVIDIA AI software stack, providing a turnkey environment for machine learning, analytics, and AI experimentation. The platform is also designed to support interconnected AI workflows, enabling organizations to scale compute performance as AI requirements continue to grow.
Enabling the future of enterprise AI agents
As organizations increasingly adopt agentic AI and autonomous AI assistants, ASUS ExpertCenter Pro ET900N G3 is engineered to support secure, local AI execution for enterprise environments. By enabling on-premises AI workflows, enterprises can maintain stronger governance over sensitive data while accelerating real-time AI operations and reducing dependency on cloud-based infrastructure. ASUS engineering teams conducted extended stress testing using vLLM, demonstrating the exceptional AI computing capabilities of ET900N G3. When running the massive Qwen open-source AI model, the system achieved approximately 864 tokens per second output throughput, with combined input and output processing reaching around 1,600 tokens per second. These results highlight ET900N G3’s ability to efficiently handle demanding large-scale open-source AI workloads and accelerate advanced AI development and inferencing at the deskside.
ET900N G3 also supports NVIDIA NemoClaw workflows, enabling enterprises and developers to build and deploy always-on AI assistants and autonomous AI agents within more secure local environments. Combined with the NVIDIA AI software stack, the NemoClaw blueprint helps simplify the development of advanced AI applications while maintaining enterprise-grade control, privacy, and operational flexibility.
The combination of powerful AI compute, extensive coherent memory, and enterprise-ready architecture makes ET900N G3 ideal for AI research labs, enterprise AI deployment, content creation, simulation workloads, and next-generation autonomous AI applications.
Subscribe for updates here
Request for pre-sales consultation here
About ASUS
ASUS is a global technology leader that provides the world’s most innovative and intuitive devices, components, and solutions to deliver incredible experiences that enhance the lives of people everywhere. With its team of 5,000 in-house R&D experts, ASUS is world-renowned for continuously reimagining today’s technologies for tomorrow, garners more than 11 awards every day for quality, innovation, and design, and is ranked among Fortune’s World’s Most Admired Companies.
View original content to download multimedia:https://www.prnewswire.com/news-releases/asus-launches-expertcenter-pro-et900n-g3-built-on-nvidia-dgx-station-architecture-302808405.html
SOURCE ASUS Computer International
