A Highlight Reel of NVIDIA CEO Jensen Huang’s GTC2026 Taipei Keynote


At the NVIDIA GTC Taipei conference held during COMPUTEX, developers, researchers, and industry leaders from around the world gathered to delve into groundbreaking innovations that are reshaping industries, spanning AI-powered factories and the expansion of infrastructure to… Proxy AI and topics such as physics-based AI.

 

Upon arriving in Taipei, NVIDIA Founder and CEO Jensen Huang has been on a nonstop schedule.

 

This week, wherever Jensen Huang went, the ecosystem was ever-present. Today, they gathered at Taipei’s Pop Music Center to attend the keynote address. Huang expressed his gratitude to everyone in the room—ranging from CEOs of various companies to the fruit vendor he happened to meet at a night market.

 

Jensen Huang told the audience both on-site and online, “Today, AI has become a profit engine—and a GDP engine as well.”

 

“Practical AI has arrived.” In the first few months of 2026, the number of commits submitted by developers on platforms like GitHub nearly tripled compared to previous levels.

 

Jensen Huang said this also makes those at the heart of this wave more valuable than ever before.

 

AI factories have become a new type of infrastructure.

 

Jensen Huang stated that tokens have now become profit-generating revenue units, and AI companies are racing to build more AI factories, driving regional demand for computing power to new heights.

 

Jensen Huang emphasized that in the AI factory era, computing power is revenue—every token generated can drive profit. This makes performance per watt, reliability, and the system’s long lifecycle key financial levers—rather than mere technical metrics: “If you have a billion watts of power, your revenue is determined by throughput per watt. Choosing the wrong architecture simply because a chip is cheaper makes no sense, since computing power drives revenue. The more you buy, the more you earn.”

 

NVIDIA’s Vera Rubin has entered full-scale mass production.

 

Subsequently, Jensen Huang announced that NVIDIA’s Vera Rubin is accelerating its transition into full-scale mass production.

 

 

He stated, “The supply chain we’ve built for Vera Rubin is twice the size of Grace Blackwell’s. We need all our production capacity to meet current demand.”

 

Agents now have their own dedicated runtime.

 

Jensen Huang views intelligent agents as the next major computing opportunity.

 

He pointed out that this shift will give rise to an entirely new CPU market “unprecedented in scale,” driven by autonomous agents that operate continuously and constantly orchestrate tools and data.

 

Jensen Huang stated, “This application model will become the computing paradigm that every company adopts over the next decade,” adding that intelligent agents will form the foundational layer of their infrastructure.

 

 

Reinventing the Computer

 

After four decades of development, personal computing has reached where it stands today. NVIDIA and Microsoft are reimagining the PC for the era of the personal intelligent agent—extending its reach from the data center all the way to the desktop.

 

Jensen Huang unveiled NVIDIA RTX Spark, enabling thin-and-light Windows laptops and compact desktop systems to deliver 1 petaflop of AI performance.

 

by MediaTek Jointly developed and powered by the Microsoft Windows operating system, NVIDIA RTX Spark will drive the first PC designed specifically for personal AI agents—always online and always running locally.

 

 

Jensen Huang said, “This is the beginning of a new journey—a completely new product line and a fresh start.” He noted that going forward, every generation of NVIDIA architecture will span desktops, laptops, and workstations, emphasizing, “The global PC industry has fully joined us in reshaping the PC.”

 

AI Enters the Physical World

 

AI is making its way into factories, vehicles, hospitals, and the myriad physical systems that keep our world running. Jensen Huang describes this as the frontier of physical AI—where intelligent agents no longer merely read and write text, but instead perceive, reason, and act in the real world.

 

 

Updates to solutions such as NVIDIA Cosmos 3, a multimodal model; Alpamayo 2 Super, an autonomous driving inference model; and the NVIDIA Isaac GR00T reference humanoid robot are helping enterprises and developers accelerate the deployment of physics‑based AI applications.

 

The full-stack product has been officially delivered.

 

Jensen Huang stated that Vera Rubin has now entered full-scale mass production. It is not merely a GPU, but a complete distributed intelligent‑agent processing system. While Grace Blackwell is primarily designed for AI computing, especially inference workloads, Vera Rubin is specifically optimized to meet the operational requirements of intelligent agents.


He stated that NVIDIA has now transformed into an infrastructure company, no longer merely a graphics‑card or systems vendor, but rather an infrastructure service provider dedicated to helping customers rapidly maximize revenue and profitability.

 

“I would like to express my gratitude for the cooperation and deep camaraderie we have shared. Without our collective efforts, today’s achievements would not have been possible.”

 

The above is the abstract; click “Read the original text” or Scan the QR code below. Read the full content:

Recommended News


GTC25 | The Chinese-subtitled version of NVIDIA CEO Jensen Huang’s keynote is now available!

Watch NVIDIA CEO Jensen Huang unveil the latest developments in AI and accelerated computing at GTC 2025.


Tsinghua Unigroup Digital joins forces with VAST Data to usher in a new era of AI storage.

Ziguang Xiaotong Technology Co., Ltd. has officially become the exclusive distributor of VAST Data in China.


Why do the world’s leading autonomous driving companies all adopt the VAST Data platform?

In the fiercely competitive autonomous‑driving industry, companies that stand out are not merely refining their algorithms—they are reimagining the data platforms that underpin connected autonomous vehicles and their broader ecosystem.


Gathering Potential Breakthrough and Sailing, Ziguang Digital Joins Hands with NVIDIA to Accelerate the Layout of the Next Station of Artificial Intelligence Development

In October 2022, Ziguang Xiaotong, a subsidiary of Ziguang Digital Group, signed an agency agreement with NVIDIA. Ziguang Digital officially became the general agent of NVIDIA Data Center GPU,NVIDIA DGX System, NVIDIA Professional Visualization, NVIDIA Virtual GPU and NVIDIA Network Products in China. Start a new chapter of cooperation with NVIDIA in the field of artificial intelligence.


Hot out of the circle! ChatGPT brings an industrial whirlwind, Ziguang Digital specially launched NVIDIA AI computing product matrix.

ChatGPT is an AI dialogue model developed by OpenAI in the United States in November 2022. Due to its excellent text generation and dialogue interaction capabilities, it has rapidly become popular worldwide, with more than one million active users in 5 days and more than 100 million active users in 2 months.