ASUS has unveiled its latest innovation, the ExpertCenter Pro ET900N G3, a powerful AI supercomputer designed to revolutionize the way enterprises, developers, and AI enthusiasts approach artificial intelligence. This deskside machine is a game-changer, bringing the capabilities of data-center-class AI directly to users' desktops, and it's an exciting development in the world of AI computing.
A Desktop Supercomputer
What makes the ET900N G3 truly remarkable is its ability to pack data-center-class AI performance into a compact deskside form factor. With the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip at its core, this machine delivers multi-petaflop-scale AI computing, a feat typically associated with large, expensive data centers. This is a significant leap forward, as it means that advanced AI development and inference can now be conducted right on one's desk.
The 748GB of coherent CPU-GPU memory is a game-changer for developers and researchers. It allows them to work with larger AI models locally, reducing the need for cloud computing and associated latency issues. This is particularly fascinating because it empowers individuals to experiment with cutting-edge AI models, such as those with over a trillion parameters, right on their local machines.
Enterprise-Grade AI Infrastructure
ASUS has cleverly designed the ET900N G3 to be enterprise-ready, built on the NVIDIA DGX Station GB300 architecture. This means that businesses and research institutions can deploy AI infrastructure without the need for dedicated data centers. The system offers enterprise-class scalability, connectivity, and reliability, making it an attractive solution for organizations looking to embrace AI without significant infrastructure overhauls.
The optimization for local AI development is a key selling point. With support for the NVIDIA AI software stack, the ET900N G3 provides a turnkey environment for machine learning, analytics, and AI experimentation. This is a significant advantage, as it streamlines the process of getting started with AI, making it more accessible to a wider range of users.
Enabling the Future of Enterprise AI Agents
One of the most intriguing aspects of the ET900N G3 is its support for agentic AI and autonomous AI assistants. By enabling on-premises AI workflows, enterprises can maintain stronger control over sensitive data while accelerating real-time AI operations. This is a critical development, as it addresses the growing demand for secure, local AI execution in enterprise environments.
ASUS's engineering teams conducted extensive stress testing, and the results are impressive. When running the massive Qwen open-source AI model, the ET900N G3 achieved an output throughput of approximately 864 tokens per second, with combined input and output processing reaching around 1,600 tokens per second. This demonstrates the system's ability to handle demanding large-scale open-source AI workloads efficiently.
A Versatile Workhorse
The combination of powerful AI compute, extensive coherent memory, and enterprise-ready architecture makes the ET900N G3 a versatile workhorse. It is well-suited for AI research labs, enterprise AI deployment, content creation, simulation workloads, and next-generation autonomous AI applications. This versatility is a significant advantage, as it means that the machine can be adapted to a wide range of use cases, making it a valuable investment for organizations looking to future-proof their AI capabilities.
In my opinion, the ET900N G3 is a game-changer for AI computing. It brings the power of data-center-class AI to the desktop, empowering individuals and organizations to experiment, innovate, and create with cutting-edge technology. As AI continues to transform industries, machines like this will play a pivotal role in driving innovation and shaping the future of AI-powered applications.
.jpg)
