Leading AI and Robotics Innovations
“Discover how InnoVISION CEO Huang Jen-hsun is spearheading a technological revolution with groundbreaking advancements in AI and robotics through the latest chip innovations.”
In an exciting keynote address titled “Unveiling the New Industrial Revolution,” NVIDIA CEO Jensen Huang revealed the company’s most recent chip designs and technological advancements on the eve of Computex 2024. In addition to summarizing NVIDIA’s outstanding accomplishments in the field of artificial intelligence, Huang also unveiled the company’s chip roadmap for the upcoming years and thoroughly examined the most recent developments in digital twin technology and industrial robots.
Huang Renxun made the initial announcement in his address that Nvidia’s new Blackwell architecture is now in production and that the company intends to keep improving its AI accelerators over the next few years. He disclosed that, in order to further solidify its dominant position in the AI space, Nvidia plans to introduce the Blackwell Ultra chip in 2025. Nvidia plans to release the Rubin chip in 2026, which will leverage cutting-edge HBM4 memory technology to boost AI applications’ performance even more. Nvidia also intends to introduce Rubin Ultra in 2027. With a data transmission speed of up to 3600 GB/s, this chip will integrate a new generation of Arm-based Vera CPU and NVLink 6 Switch, significantly increasing the processing capability of data centers.
In his address, Huang Renxun emphasized that the Blackwell chip is “the most complex and high-performance computer in the world” and that, over the previous eight years, its AI computing capacity has increased 1,000 times while consuming only 1/350 of the energy. He used the example of how training a GPT-4 model with 1.8 trillion parameters and 8 trillion tokens only requires 3 GWh of energy, which is significantly less than the 1,000 GWh of the Pascal architecture from eight years ago. Furthermore, Blackwell has lowered the energy usage for token generation to 1/45,000 of what it was eight years ago. This development will significantly encourage the use of AI in many different industries.
Huang Renxun first unveiled the chip strategy before showcasing NVIDIA’s most recent advancements in robotics and digital twin technologies. According to him, businesses may create a smooth transition between the virtual and physical worlds and discover innovative approaches to product design, manufacturing, and testing with the use of NVIDIA’s digital twin technology. Simultaneously, NVIDIA is actively investigating the use of robots in diverse industries, which has produced impressive outcomes.
Huang Renxun concluded by saying that NVIDIA would remain dedicated to advancing the advancement of robots and AI technologies and that the company aspires to collaborate with international partners to build a brighter future. He thinks that when technology develops further and the range of applications grows, artificial intelligence (AI) and robotics will play a significant role in advancing societal progress.
Are you ready to dive deeper into the topics you love? Visit our website and discover a treasure trove of articles, tips, and insights tailored just for you!
