AI Infrastructure

FAQs

• What is an AI server?

  An AI server is a computer system specifically designed to power AI; but unlike standard office servers, they're able to crunch through the incredibly complex calculations that fuel AI advancements.
• What is the ASUS AI Factory, and how does it enable organizations to scale AI solutions?

  The ASUS AI Factory is built for token generation. It is a comprehensive, end-to-end platform that transforms raw data into actionable intelligence, enabling organizations to scale ideas into real-world AI solutions. By integrating advanced hardware, software, and services, ASUS delivers a holistic AI ecosystem—supporting both cloud-based and on-premises deployments—for a wide range of enterprise AI workloads.
• How does ASUS deliver advanced AI infrastructure and solutions?

  ASUS partners closely with NVIDIA to deliver full-stack, token-generated AI Factory solutions that integrate validated hardware platforms (such as NVIDIA Blackwell, HGX, and MGX architectures) with enterprise-grade software, networking, and storage. This collaboration ensures optimized performance, accelerated AI adoption, reduced total cost of ownership (TCO), and seamless deployment of agentic and physical AI solutions.
• What is the role of a desktop AI supercomputer such as the Ascent GX10 within the broader AI infrastructure?

  The Ascent GX10 serves as a powerful local AI development and deployment hub. Its primary roles are: Prototyping & Fine-Tuning: It handles large parameter models (up to 200B) directly on the desktop, freeing up large, expensive cloud or data center clusters for final training and production deployment. Edge & Local Inference: It enables high-performance, real-time AI inference at the edge, providing immediate insights for applications in manufacturing, retail, and smart cities, where low latency is critical. Decentralized Security: It allows organizations to keep sensitive data and models local and confidential, bypassing the privacy and security risks associated with public cloud services.
• How does the Ascent GX10 achieve supercomputer-level AI performance in such a compact, desktop-friendly design?

  Performance of the Ascent GX10 is driven by the state-of-the-art NVIDIA GB10 Grace Blackwell Superchip. This architecture results in multiple benefits. •Integration of the CPU and GPU onto a single platform with 128GB of Coherent Unified System Memory eliminates data-transfer bottlenecks common in traditional systems. •Up to 1 petaflop of AI performance (FP4) is readily available―a level previously reserved for large-scale data centers. •Efficient design and advanced thermals ensure sustained high performance in a compact, office-friendly form factor.
• Can the Ascent GX10 be scaled for larger models and expanding AI demands?

  Yes, the Ascent GX10 offers a scalable architecture to meet growing AI demands. By combining two ASUS Ascent GX10 units, users can significantly expand AI capacity. This configuration doubles the Unified Memory from 128GB to 256GB, enabling support for AI models of up to 405B parameters. Storage capacity also expands to as much as 8TB. This provides a smarter way to scale demanding AI workloads.
• What is an AI Supercomputer?

  An AI supercomputer is a high-performance computing system specifically engineered to train, fine-tune, and deploy advanced AI models at scale. Unlike traditional desktops or workstations, AI supercomputers integrate extreme compute density, GPU-accelerated performance, high-bandwidth memory, and ultra-fast interconnect technologies to handle massive datasets and complex AI workloads such as generative AI, large language models (LLMs), scientific simulations, and real-time analytics. The ASUS ExpertCenter Pro ET900N G3 represents the next generation of desktop AI supercomputing, powered by the NVIDIA GB300 Grace Blackwell Ultra Desktop Superchip. By combining CPU, GPU, memory, and ultra-fast interconnect into a unified architecture, ET900N G3 delivers breakthrough performance for enterprise-level AI workloads—including LLM training, simulation, and multi-modal AI development—within a deskside form factor. In short, an AI supercomputer delivers extreme AI performance once only found in data centers, and the ET900N G3 brings that capability directly to offices, labs, and research environments.