Understanding Huang's ComputeX Speech in One Article: This is Not a Product Launch, This is the "AI Industrial Revolution Mobilization Order"

On May 19, 2025, Jensen Huang, founder and CEO of NVIDIA, delivered a two-hour keynote speech at the Computex 2025 conference.

From AI infrastructure, chip platforms, enterprise AI, to robotics and digital twins... Huang painted a picture of a rising AI Factory era.

In the past, data centers served traditional applications; today’s AI data centers are no longer just “data centers”, but AI factories: a new type of intelligent factory that inputs electricity and outputs “Tokens.”

“NVIDIA is no longer just a technology company; we are now an AI infrastructure company.”

He emphasized that this is the third infrastructure revolution following electricity and the internet—intelligent infrastructure.

Major Chip Release: Grace Blackwell GB200 and NVLink Architecture

Huang showcased the NVLink Spine—a core interconnect module weighing 70 pounds, featuring two miles of cabling and 130 TB/s bandwidth. He stated, “The data throughput of this system is larger than the entire internet!”

The GB200 Grace Blackwell superchip uses a dual-chip package and connects 72 GPUs, resembling a “virtual giant chip.” It is built on the latest NVLink Spine architecture, with a single node equivalent to the performance of the 2018 Sierra supercomputer.

Huang stated:

“This is not a server; this is an AI factory. You input energy into it, and it will give you Tokens.”

Additionally, the event also announced the upgrade of the GB200 system. NVIDIA plans to launch the GB300 in Q3, which will enhance inference performance by 1.5 times, increase HBM memory by 1.5 times, and double network bandwidth, while maintaining physical compatibility with the previous generation to achieve 100% liquid cooling.

NVLink Fusion: Open Chip Interconnect Ecosystem

The most eye-catching aspect is the NVLink Fusion plan.

The NVLink Fusion architecture allows seamless connection of CPUs / ASICs / TPUs from other manufacturers with NVIDIA GPUs. This technology provides NVLink Chiplet & interface IP, allowing for the free combination of “semi-custom infrastructure.” **

In simple terms, customers can choose to use their own CPUs alongside NVIDIA's AI chips, or use NVIDIA's CPUs with AI accelerators from other vendors.

Some analysts believe that NVLink, as one of the key technologies for NVIDIA to ensure its dominance in AI workloads, addresses the communication speed issue between GPUs and CPUs in AI servers, which is one of the biggest barriers to scalability and directly affects peak performance and energy efficiency. It offers higher bandwidth and lower latency compared to standard PCIe interfaces, with bandwidth advantages of up to 14 times.

NVLink Fusion allows Fujitsu and Qualcomm to use this interface for their own CPUs, with NVLink functionality integrated into the chipset next to the computing package. NVIDIA has also attracted custom silicon accelerators from designers such as MediaTek, Marvell, and Alchip, supporting other types of custom AI accelerators to work in conjunction with NVIDIA's Grace CPU.

Jensen Huang humorously said:

"Of course, it's best if you use all NVIDIA, that would make me the happiest. But if you only use a little bit of NVIDIA, I'm still very happy."

Personal Supercomputing Era: DGX Spark and DGX Station

Jensen Huang stated that the personal AI computer DGX Spark, previously mentioned at CES as Project DIGITS, has now fully entered production and will be launched in the coming weeks.

DGX Spark is aimed at AI researchers who want to own their own supercomputers, with companies setting their own prices. Jensen Huang remarked, "Everyone can have one by Christmas."

"Today, everyone can own their own AI supercomputer, and... it can plug into a kitchen outlet."

Enterprise AI Restructuring: From Hardware to Agentic AI

Jensen Huang also announced the RTX Pro Enterprise AI Server: supporting traditional x86, Hypervisor, Windows, and other IT workloads. It can also run graphical AI agents, even capable of running the Crysis game.

Jensen Huang stated that Agentic AI is the future's "digital employee." Digital customer service representatives, digital marketing managers, digital engineers, etc., will become part of the corporate workforce. NVIDIA will deploy AI Ops full-stack support and collaborate with CrowdStrike, Red Hat, DataRobot, and others to promote enterprise AI deployment.

"We will need new HR to manage these AI employees."

Brand New AI Storage Architecture: NVIDIA AIQ + Nemo + GPU Storage Frontend

Jensen Huang stated that AI no longer just reads SQL; it needs to understand the semantics of unstructured data. Future storage systems will have built-in GPUs for searching, sorting, embedding, and indexing.

NVIDIA will also deploy Nemo + NeMo Retriever + IQ, an open-source "AI semantic retrieval framework," and collaborate with Dell, Hitachi, IBM, NetApp, and VAST to deploy enterprise-level platforms.

Robots Will Become the Next "Trillion-Dollar Industry"

Jensen Huang indicated that they are advancing robotic systems in parallel with the automotive industry, using the Isaac Groot platform, which is powered by a new processor called Jetson Thor, designed specifically for robotic applications, suitable for everything from autonomous vehicles to human-machine systems. NVIDIA's Isaac operating system manages all neural network processing, sensor processing, and data pipelines, leveraging pre-trained models developed by a specialized robotics team to enhance system capabilities.

"In the AI era, to train robots... you first have to use AI to teach AI."

Jensen Huang also mentioned that they are applying their AI models to autonomous vehicles, launching a fleet globally in collaboration with Mercedes, utilizing NVIDIA's end-to-end autonomous driving technology, which can be realized this year.

He believes that robots will become the next industry worth trillions of dollars, but this requires a lot of effort. NVIDIA's robotics division is capable of achieving this, and this is solely due to scalability.

Launching the Physical AI Engine Newton

Jensen Huang stated that they have collaborated with DeepMind and Disney Research to develop the world's most advanced physics engine, Newton, which is planned to be open-sourced in July.

According to reports, Newton fully supports GPU acceleration, has high differentiability and ultra-real-time operational capabilities, and can achieve effective learning through experience. NVIDIA is integrating this physics engine into NVIDIA's ISAAC simulator, enabling these robots to "come to life" in a realistic manner