Scan AI

Get in touch with our AI team.

AI at Computex

Find out about the latest innovations in AI at the Computex trade show in Taiwan



The Computex trade show takes place annually in the heart of the computer industry, Taiwan. See all the highlights of NVIDIA CEO Jensen Huang’s keynote from the show.

Get in touch with the Scan AI team about anything regarding AI.


Be the first to know

Discover the next-gen Blackwell GPU architecture and AI appliances below and register your interest to be the first in the queue when more information is available.


GB200 NVL72

After teasing the DGX GB200 SuperPOD at GTC March 2024, NVIDIA has released further details and changed the name to GB200 NVL72. Designed for maximum density, GB200 NVL comprises 36 Grace CPUs and 72 Blackwell GPUs in a single rack, scalable as large as you need.


Future Architectures

After announcing the Blackwell GPU architecture announced at GTC March 2024, NVIDIA announced this will be followed by an improved version Blackwell Ultra, planned for launch in 2025. NVIDIA also teased the Rubin GPU and Vera CPU architectures, which are planned for 2026.


Future Networking

AI systems need rapid access to data so NVIDIA announced an expansion of its ecosystem of 800Gb/s networking products planned for launch in late 2024. NVIDIA also announced it is planning to ship 1,600Gb/s networking products, including CX9 SuperNICs and X1600 converged InfiniBand/Ethernet switches in 2025.



NIM is a collection of pre-built containers that speed up deploying generative AI projects. Announced at GTC March 2024 NIM is now supported by 200 partners for domain-specific applications, such as copilots, code assistants, digital human avatars and more.


GeForce RTX AI PCs

NVIDIA also announced that RTX AI Toolkit will help deliver more than 500 AI-powered applications and games to PCs featuring GeForce RTX GPUs. RTX AI PCs will also be able to run NVIDIA frameworks including ACE, Audio2Face and Riva.