NVIDIA unveils breakthroughs at COMPUTEX 2024

Key announcements included the introduction of the Grace Hopper Superchip, the launch of the DGX GH200 AI supercomputer, and new software platforms such as NVIDIA AI Foundations and NVIDIA Omniverse.

Grace Hopper Superchip and DGX GH200 AI Supercomputer

The Grace Hopper Superchip combines the versatility of a CPU with the power of a GPU, aimed at accelerating AI and high-performance computing tasks.

Huang stated: “Our new Grace Hopper Superchip is designed to handle the most demanding AI workloads, from language models to recommendation systems.”

The DGX GH200 AI supercomputer, leveraging this technology, was designed specifically for generative AI and large language models.

Huang highlighted its capabilities: “The DGX GH200 will enable researchers and companies to push the boundaries of what’s possible in AI.”

New software platforms

NVIDIA AI Foundations offers a suite of tools for developing and deploying AI applications, while NVIDIA Omniverse aims to revolutionise industrial digitalisation by creating a unified platform for 3D design and simulation.

Huang mentioned: “NVIDIA AI Foundations provides the building blocks for creating next-generation AI solutions, and Omniverse is transforming how industries approach digitalisation.”

Strategic partnerships

NVIDIA announced collaborations with tech giants like Microsoft, Google, and Meta to advance AI and Cloud computing. These partnerships aim to integrate NVIDIA’s AI technology into their Cloud services, enhancing their computational power and efficiency.

Huang emphasised NVIDIA’s commitment to innovation: “By partnering with industry leaders, we are ensuring that our technology reaches a broader audience and drives the next wave of AI breakthroughs.”

In summary, Huang’s keynote underscored NVIDIA’s leadership in AI and high-performance computing, with new hardware and software innovations poised to drive future advancements in the field.

For more details, you can watch the keynote here.