NVIDIA and ASUS Build AI Supercomputer for Taiwan's Research

NVIDIA and ASUS are collaborating to build a new AI supercomputer at Taiwan's National Center for High-Performance Computing (NCHC), announced in a company blog post. This supercomputer is expected to deliver over eight times the AI performance of the center's previous system, Taiwania 2.

The new system will feature NVIDIA HGX H200 systems with over 1,700 GPUs, two NVIDIA GB200 NVL72 rack-scale systems, and an NVIDIA HGX B300 system, all interconnected by NVIDIA Quantum InfiniBand networking. The supercomputer is set to go live later this year and will support a range of research areas, including climate science, quantum computing, and the development of large language models.

NCHC plans to deploy additional NVIDIA DGX Spark personal AI supercomputers and a cluster of NVIDIA HGX systems in the cloud. Researchers from academic institutions, government agencies, and small businesses in Taiwan will have the opportunity to apply for access to this advanced computing resource to accelerate their innovative projects.

The supercomputer will also support the Taiwan AI RAP platform, which aids in the rapid development of AI products by providing access to customized models that reflect local cultural and linguistic nuances. This initiative includes the Trustworthy AI Dialogue Engine (TAIDE), which is focused on building Taiwanese large language models for various applications, including natural language processing and intelligent customer service.