At GTC 2026 in San Jose, enterprise AI was deployed across data centers, edge sites, and physical systems. Real business value depends on infrastructure, software, and services that are reliable, scalable, and practical—a complete ecosystem is essential for AI to deliver results. ASUS combines powerful hardware, flexible infrastructure, and expert services to help organizations put AI to work from data centers to the edge.
Building on these initiatives, ASUS recently announced a strategic expansion of its Infrastructure Solutions Business Group (IS BG), reinforcing its commitment to next-gen AI infrastructure. The IS BG will deliver integrated, future-ready systems—including servers, storage, networking, and edge-to-cloud solutions—while adding R&D, sales, and operational resources to support high-growth customers worldwide.
This year at GTC, ASUS demonstrated our latest innovations, highlighting how the right ecosystem of infrastructure, software, and services drives tangible results.
ASUS AI POD built on NVIDIA Vera Rubin NVL72 platform
ASUS unveiled its fully liquid-cooled AI infrastructure at NVIDIA GTC 2026
Scaling AI at the Enterprise Level
ASUS showcased next-gen AI infrastructure designed for large-scale deployments—from rack-scale AI factories to enterprise and edge environments. The ASUS AI POD, powered by NVIDIA Vera Rubin NVL72, delivers ultra-dense AI performance, flexible cooling, enterprise-grade redundancy, and simplified full-stack deployment.
Our portfolio spans liquid-, hybrid-, and air-cooled systems, giving organizations the flexibility to build energy-efficient clusters tailored to their needs. Through close collaboration with NVIDIA, we are preparing the AI POD for mass production and engaging with customers to meet anticipated demand. These solutions allow enterprises to scale AI safely and efficiently while maintaining operational control and energy efficiency.
Rack-scale servers such as XA NR1I-E12L/R, and high-performance systems including XA NB3I-E12, ESC8000A-E13X, and ESC8000A-E13P support workloads ranging from AI model training to high-end visual computing. Additionally, our partnerships with Vertiv, Schneider, and NVIDIA-certified storage providers ensure resilient, optimized, and redundant performance across every deployment.
ASUS server with NVIDIA HGX B300 systems
ASUS MGX Server with NVIDIA RTX PRO 4500 Blackwell Server Edition or NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs
ASUS AI Storage Solutions
Bringing AI into the Physical World
ASUS is also extending AI beyond the data center, powering robotics, automation, and intelligent systems. Some of our key solutions in this area include:
PE3000N – Rugged edge AI platform powered by NVIDIA Jetson Thor, designed for real‑time compute, sensor fusion, and autonomous operations.
On the software side, ASUS AI Hub provides a turnkey platform for deploying agentic AI, building custom AI assistants, implementing RAG-enhanced document intelligence, and scaling secure operations. Together, these systems allow AI to perceive, reason, and act, bridging the digital and physical worlds and turning experimentation into results.
ASUS Ascent GX10
ASUS ExpertCenter Pro ET900N G3
ASUS PE3000N is edge AI computer powered by NVIDIA Jetson Thor™ for real-time AI inference in robotics, autonomous vehicles, and intelligent video analytics
Key GTC Takeaways
During the event, my colleagues from ASUS took the stage to share how enterprise AI can be operationalized at scale. Their sessions highlighted how infrastructure, software, and orchestration come together to enable real-world AI deployments—from data centers to edge sites, and from digital models to physical systems. These presentations offered a clear view of how ASUS partners with NVIDIA, IBM, and other ecosystem leaders to deliver solutions that turn AI potential into measurable outcomes.
Building the Enterprise AI Factory
In the session “Building the Enterprise AI Factory: A Unified Stack with ASUS, IBM, and NVIDIA,” my colleagues Alber Wu, ASUS Division Director of Solutions, and Vincent Hsu (VP, IBM Storage Solutions) shared how enterprises can turn raw data into actionable insights at scale. They highlighted a converged, validated AI architecture enabling reliable and efficient deployment from the data center to the edge:
AI Factories for the Generative AI Era: Enterprises must transform raw data into actionable insights at scale. ASUS, IBM, and NVIDIA provide a converged “AI refinery”, enabling secure, scalable Agentic AI deployment from data centers to the edge.
Integrated Converged Architecture: The solution combines NVIDIA-certified ASUS GPU servers, NVIDIA AI Enterprise pipelines, and IBM Storage with embedded intelligence—supporting vector indexing, real-time change detection, secure access, KV cache offload, and optimized inferencing.
Orchestration & Hybrid Deployment: Red Hat OpenShift, integrated with NVIDIA GPU Operator, NVIDIA NIM Operator, and IBM Content-Aware Storage (CAS), enables secure, containerized deployments across hybrid environments, supporting advanced inferencing and Retrieval-Augmented Generation (RAG) workflows.
Validated AI Blueprints: ASUS GPU servers and AI blueprints accelerate deployment, ensuring trust, performance, and scalability for enterprise-grade AI workloads.
Centralized Management & Deployment: ASUS Infrastructure Deployment Center (AIDC) enables rapid NVIDIA-certified storage deployment, while ASUS Control Center (ACC) provides real-time monitoring of GPU servers, storage servers, and CAS/Spectrum Scale SDS service health.
AI Factory in Action — Achieving Agile Automation with Gen AI Agents and Digital Twins
In “AI Factory in Action: Achieving Agile Automation with Gen AI Agents and Digital Twins,” my colleague Joseph Lu, ASUS Senior Director, presented alongside Jesse Chen, CEO of Spingence. They explored how factories are evolving into AI Factories, where intelligence is created alongside products:
The AI Factory Paradigm: As manufacturing shifts from mass production to high-mix, low-volume demands, flexibility is critical. The AI Factory integrates intelligence into production, enabling adaptive, autonomous manufacturing.
Agentic AI in Action: AI Agents perceive changes, reason through complex tasks, and validate strategies in virtual simulations before physical execution, shortening robot workflow adaptation from weeks to minutes.
Converged Infrastructure & Edge AI: ASUS provides high-performance Edge AI hardware and NVIDIA AI Enterprise for secure, scalable model deployment, forming the backbone for autonomous operations and generative AI workflows.
Digital Twins & NVIDIA Omniverse: NVIDIA Omniverse enables physically accurate simulations, allowing manufacturers to test and optimize processes virtually before applying them on the shop floor.
Full-Stack AI-Focused Solution: ASUS’s certified AI servers and edge devices, integrated with NVIDIA AI Enterprise and Omniverse, streamline generative AI deployment and transform fragmented automation into a cohesive, intelligent system.
Collaboration at the Core
What stood out most at GTC 2026 is that enterprise AI thrives through collaboration. The combination of ASUS infrastructure, NVIDIA platforms, and partner expertise from IBM storage to Omniverse digital twins shows that complex AI deployments succeed when technology, guidance, and services come together.
Building impactful AI is never a solo effort. It requires trusted partnerships, integrated solutions, and a shared commitment to performance, scalability, and real-world outcomes. At ASUS, I’m proud to work with industry leaders and our customers to deliver AI ecosystems that turn potential into measurable results, whether in the data center, at the edge, or in the physical world.