Nvidia Introduces "Jetson Thor" Chip Module
Applications Span Robots and Autonomous Vehicles
Nvidia, an artificial intelligence (AI) semiconductor company, announced on August 25 (local time) that it will begin selling its latest robotics chip module, the "Jetson AGX Thor."
Nvidia refers to this chip as a "robot brain" and explained that developers can use it to build robots. The company also stated that after companies create robot prototypes using the developer package, they can use the "Thor T5000" module for mass production.
The "Jetson Thor" chip is based on Nvidia's Blackwell graphics processing unit (GPU), which is also used in the company's current AI chips and gaming chips, and it is 7.5 times faster than the previous generation. It is equipped with 128GB of memory, which is essential for running large-scale AI models. Nvidia explained that this enables the execution of generative AI models required for humanoid robots. The company also added that this chip can be used in autonomous vehicles.
Deepu Talla, Vice President of Robotics and Edge AI at Nvidia, stated, "We do not build robots or cars ourselves, but we enable the entire industry with infrastructure computers and the accompanying software."
Robotics is the area that Nvidia sees as its greatest growth opportunity. Chief Executive Officer Jensen Huang has previously indicated his intention to target this market, stating, "After AI, robotics could become the next largest growth market."
Although Nvidia has been releasing various robot chips since 2014, the robotics sector still accounts for only about 1% of the company's total revenue.
Nvidia revealed that robotics companies such as Agility Robotics, Boston Dynamics, Amazon, and Meta are using Jetson chips. Nvidia is also investing in robotics companies like Field AI.
© The Asia Business Daily(www.asiae.co.kr). All rights reserved.
