🎉 #Gate Post# Hits 50,000 Followers!
✨ To celebrate this amazing milestone, we're giving back to our incredible community!
🎁 4 Lucky Winners Will Each Receive $10 Points!
Join:
1️⃣ Follow Gate_Post
2️⃣ Like this post
3️⃣ Drop your congratulations in the comments!
End at 18:00, May 25 (UTC)
NVIDIA collaborates with the National Center for High-Performance Computing to create a comprehensive sovereign AI for Taiwan.
NVIDIA CEO Jen-Hsun Huang emphasized Taiwan's key role in the global AI technology landscape during the Computex 2025 conference held in Nangang, Taipei. Taiwanese tech companies are helping to build an artificial intelligence ecosystem worldwide, and the Taiwanese government is also working on deploying sovereign AI infrastructure with the development of AI supercomputers. NVIDIA provides technical resources in partnership with the National Center for High-Performance Computing to accelerate the innovation of Taiwan's local AI technology.
NVIDIA announced its partnership with Foxconn (Hon Hai Technology Group) to build the Taiwan AI Factory supercomputer, which will provide cutting-edge NVIDIA Blackwell infrastructure for startups, researchers, and companies including TSMC.
NVIDIA collaborates with the National Center for High-Performance Computing to create sovereign AI
Researchers in Taiwan are facing complex technological challenges in developing artificial intelligence, climate science, and quantum computing. They will soon receive assistance from Taiwan’s National Center for High-Performance Computing, abbreviated as NCHC), with its new supercomputer. The artificial intelligence performance of this center will be over 8 times higher than that of the previously launched Taiwania 2 system.
The AI supercomputer project of NCHC utilizes the NVIDIA HGX H 200 system with over 1,700 GPUs, two NVIDIA GB 200 NVL 72, and an NVIDIA HGX B 300 system built on the NVIDIA Blackwell Ultra platform, connected via NVIDIA Quantum InfiniBand Networking. This functionality is scheduled to go live later this year.
The NCHC also plans to deploy a set of NVIDIA DGX Spark personal AI supercomputers and a NVIDIA HGX system cluster in the cloud. Researchers from academic institutions, government agencies, and small businesses in Taiwan will be able to apply to use the new system to accelerate innovative projects.
Zhang Chaoliang, director of the National Center for High-Performance Computing in Taiwan, stated that the new supercomputer will promote sovereign AI, quantum computing, and advanced scientific computing, enhancing Taiwan's technological autonomy and fostering interdisciplinary collaboration and global leadership in artificial intelligence.
Developing Taiwan's Sovereign AI Language Model
The new supercomputer will support the Taiwan AI RAP project, which is a generative AI application development platform. Taiwan AI RAP provides customized language models that are specifically designed for local culture and the nuances of language.
The language models provided by the platform include the Taiwan Artificial Intelligence Dialogue Engine (TAIDE) model. TAIDE is a public program aimed at processing natural language to establish a large language model (LLM) localized to Taiwan for AI clients and translation. This program offers collaboration with partners providing text, images, audio, and video data, including public sectors such as local governments, news agencies, the Ministry of Education, and the Ministry of Culture.
To support sovereign AI application development, TAIDE currently provides developers with a series of Llama 3.1-TAIDE foundational models, and the team is building additional sovereign AI LLM services using the NVIDIA Nemotron model.
Taiwanese language artificial intelligence robot activation, shortening the time for教材 development.
A university professor in Tainan uses the TAIDE model to drive an artificial intelligence robot that can communicate in Taiwanese and English with elementary and middle school students. So far, more than 2,000 students, teachers, and parents have used it. Additionally, another professor utilizes this model to generate teaching materials, shortening the time teachers need to prepare lessons.
AI healthcare application scenarios
In the healthcare field, a research team in Taiwan has developed an AI chatbot with enhanced indexing generation capabilities using the TAIDE model, which can assist medical record managers in providing timely and accurate medical information to patients with serious illnesses. The Taiwan Centers for Disease Control's Epidemic Prevention Center is training the model to generate news summaries to support the tracking and prevention of disease transmission.
NVIDIA Earth-2 platform accelerates the development of climate science research
In climate research, NCHC supports researchers in using the NVIDIA Earth-2 platform to advance the development of atmospheric science. These researchers utilize Earth-2's CorrDiff AI model to improve the accuracy of weather models and employ DeepMind's GraphCast model in NVIDIA PhysicsNeMo for global weather forecasting.
NCHC also adopted NVIDIA NIM microservices for FourCastNet, which is an NVIDIA model for predicting global atmospheric dynamics for weather and climate. It utilizes NVIDIA GPUs to accelerate the numerical simulations of weather forecasting models. With the new supercomputer, researchers will be able to run more complex simulations and speed up AI training and inference.
NVIDIA CUDA-Q and cuQuantum drive quantum research
NCHC researchers are advancing quantum computer research using the NVIDIA CUDA-Q platform and NVIDIA cuQuantum, applying it to quantum machine learning, chemistry, finance, cryptography, and other fields. This research utilizes quantum circuits and a quantum molecular generator developed on the CUDA-Q platform, which is a tool for generating effective chemical molecules (Chemical Molecules). They have also created cuTN-QSVM, an open-source tool built on cuQuantum, which can accelerate large-scale quantum circuit simulations. This tool enables researchers to tackle more complex problems, provides linear scalability, and supports hybrid quantum computing systems to help accelerate the development of large-scale quantum algorithms.
NCHC researchers recently conducted a simulation of 784 qubits using cuTN-QSVM for quantum machine learning algorithms. The research institute also plans to build a hybrid quantum accelerated computing system by integrating the NVIDIA DGX Quantum system.
This article about NVIDIA partnering with the National Center for High-Performance Computing to create a comprehensive sovereign AI for Taiwan first appeared in Chain News ABMedia.