ASUS Revolutionizes Local AI Computing with the ET900N G3
LAS VEGAS, USA — At CES 2026, ASUS made a bold statement in the world of artificial intelligence (AI) computing by unveiling the ExpertCenter Pro ET900N G3. This isn't just another workstation; it's a deskside AI supercomputer that brings data-center-level performance directly to researchers, developers, and enterprises. Powered by NVIDIA's cutting-edge Grace Blackwell Ultra platform, the ET900N G3 is poised to redefine how AI workloads are handled locally, bridging the gap between cloud-scale computing and the immediate needs of on-premise development.
The Power of a Supercomputer in a Deskside Form Factor
The ET900N G3 is built around NVIDIA's GB300 Grace Blackwell Ultra Desktop Platform, which combines a NVIDIA Grace CPU and a NVIDIA Blackwell Ultra GPU. These components are connected via NVIDIA NVLink-C2C, a high-bandwidth, low-latency chip-to-chip interconnect that enables petaflop-scale AI computing in a standalone system. This architecture is designed to handle the most demanding AI workloads, from large language model (LLM) fine-tuning to complex simulations and high-speed inference.
Massive Memory and Unprecedented Performance
A key innovation of the ET900N G3 is its 775GB of coherent unified memory. This massive memory pool, more than double that of a typical four-GPU workstation, allows developers to run larger models and accelerate experimentation without the limitations of traditional systems. The system's AI performance can reach up to 20 PFLOPS (petaflops), making it one of the most powerful deskside AI supercomputers on the market.
Designed for Next-Generation AI Workloads
The ET900N G3 is one of the first systems based on NVIDIA's DGX Station architecture to reach the deskside market. This architecture is tailored for local AI workloads, making it ideal for large-scale model training and inference. It supports a wide range of AI applications, including LLMs, advanced simulations, multi-modal AI, and deep learning research. Paired with NVIDIA's AI software stack, the ET900N G3 offers a turnkey development environment optimized for data science, machine learning, and analytics.
Scalability and Enterprise-Grade Integration
The ET900N G3 is designed to be more than just a powerful workstation; it's a system that can scale to meet the demands of modern AI development. It can be interconnected with other systems to expand its supercomputing power twofold when needed, providing flexibility and scalability for enterprises and research institutions. Additionally, its enterprise-grade software integration ensures seamless deployment and management.
A Critical Middle Ground: Power Without Complexity
One of the biggest challenges in AI development has been the 'last-mile' problem—where cloud services offer immense scale but local workflows require immediate, private compute power. The ET900N G3 solves this by providing supercomputer-level capability without the complexity of server infrastructure. This makes it accessible for businesses, research institutes, and advanced creators who need to adopt AI workflows rapidly and securely.
Unlocking New Possibilities for Innovation
As AI model sizes grow and real-time, private compute becomes essential, the ET900N G3 brings the power of a data center directly to the professionals who need it most. It unlocks new possibilities for innovation, research, and next-generation AI development. With its exceptional performance, massive coherent memory, and enterprise-grade integration, the ET900N G3 is set to become a cornerstone in the future of AI computing.
ASUS: A Leader in Innovation
ASUS has long been recognized for its commitment to innovation. With over 5,000 in-house R&D experts, the company consistently pushes the boundaries of technology, delivering solutions that enhance lives worldwide. The ET900N G3 is a testament to this commitment, representing a significant leap forward in accessible AI supercomputing. As ASUS continues to drive towards a net-zero enterprise and a circular economy, it remains dedicated to creating shared value for all stakeholders.
