ASUS announced that, in collaboration with the National Center for High-performance Computing (NCHC), a new Artificial Intelligence supercomputer is now officially in operation. The system uses a dual-compute architecture that pairs a cluster with the latest NVIDIA GB200 NVL72 system and represents Taiwan’s first fully liquid-cooled Artificial Intelligence supercomputer deployment of this architecture. The Nano4 NVIDIA HGX H200 system delivers up to 81.55 PFLOPS of performance and is ranked No. 29 on the TOP500 list, a milestone the parties say will strengthen Taiwan’s Artificial Intelligence-computing capabilities and accelerate its intelligent transformation.
NCHC served as the designer of the Nano4 and led the architecture, liquid-cooling strategy, and system integration. ASUS provided Artificial Intelligence-infrastructure expertise to complete the build and operational readiness. The deployed configuration combines the Nano4 nodes with the NVIDIA GB200 NVL72 hardware to deliver high-throughput compute for a range of workloads. According to NCHC spokesperson Dr. Chia-Lee Yang, “As the designer of the Nano4, NCHC drove the architecture, liquid-cooling strategy, and system integration. Along with Artificial Intelligence-infrastructure expertise from ASUS, this newly built system feeds advanced computes for generative Artificial Intelligence, big data, deep learning, and HPC, unlocking unprecedented computational capabilities for both academia and industry.”
The announcement frames the deployment as both a technical and strategic advance for domestic research and industry users. By combining liquid cooling with a dual-compute architecture and leading NVIDIA hardware, the installation aims to provide dense, efficient compute capacity for generative Artificial Intelligence and high-performance computing workloads. ASUS and NCHC position the online Nano4 system as a shared resource to accelerate research, model development, and application of Artificial Intelligence across academic and commercial fields in Taiwan.
