On March 19th, NVIDIA founder and CEO Huang Renxun announced in a GTC keynote speech that the centralized in car computing platform "NVIDIA DRIVE Thor" will be equipped with a new NVIDIA Blackwell architecture specifically designed for Transformers, Large Language Models (LLM), and Generative AI workloads.
This means that the AI capabilities of future cars will be further enhanced.
Several leading electric vehicle manufacturers have showcased their next-generation AI models based on NVIDIA's latest technology, including many Chinese car companies such as BYD, GAC Aion, Xiaopeng, Ideal Automobile, and Geely Krypton. Wenyuan Zhixing is also working with Lenovo Car Computing to create multiple commercial L4 level autonomous driving solutions based on DRIVE Thor.
In addition, BYD plans to use NVIDIA's AI infrastructure for cloud AI development and training, and use NVIDIA Isaac and NVIDIA Omniverse platforms to develop tools and applications for virtual factory planning and retail configurators.
"Accelerating computing has brought revolutionary breakthroughs to fields such as generative AI, which is redefining autonomous driving and the global transportation industry," said Wu Xinzhou, Vice President of NVIDIA Automotive Business Unit.
Last August, Wu Xinzhou, former Vice President of Autonomous Driving at Xiaopeng Motors, joined NVIDIA. He is considered the soul of Xiaopeng Automobile's autonomous driving business and a recognized top talent in the industry.
In 2015, Nvidia entered the field of in car computing platforms and launched the first generation autonomous driving computing platform DRIVE PX and Tegra series in car chips. Since then, the DRIVE platform has been continuously updated, and Nvidia has also been continuously upgrading its automotive grade SoC chips, and each upgrade has made significant improvements in chip performance.
In 2016, Nvidia released the Xavier chip with a computing power of 30Tops, and achieved mass production in 2020. In 2019, Orin chips were introduced with a computing power of up to 254Tops, and in 2022, they were installed on the NIO ET7 for mass production.
In terms of indicators such as deep learning acceleration, memory and communication, and CPU performance, Nvidia DRIVE Orin series chips are currently at the highest level in the market. According to Fraser Sullivan statistics, Nvidia's shipment volume accounted for 82.5% of the global market share of high computing power autonomous driving chips in 2022.
In 2022, as the growth rate of the automotive business slowed down, Nvidia released a new generation of autonomous driving computing chip DRIVE Thor. At that time, it announced a single chip with a computing power of up to 2000TOPS, which was 14 times the computing power of Tesla's FSD chip.
The special feature of NVIDIA's chips lies in their high integration. DRIVE Thor can provide intelligent cockpit, highly automated driving, and unmanned driving functions, and integrate all functions onto the same centralized platform. The host factory can use this chip to allocate tasks freely between different tasks without the need to equip products from other chip companies.
According to Huang Renxun's plan, the future automotive business will become Nvidia's three pillar businesses alongside data centers and gaming. He has publicly stated, "In the future, Nvidia's potential market size will reach $1 trillion. Specifically, in terms of segmented markets, the automotive industry's revenue will account for 33%, reaching $300 billion."
However, according to the financial report, Nvidia's total revenue for the 2024 fiscal year reached $60.9 billion, a year-on-year increase of 126%, with most of the performance surge attributed to the data center department. In the fourth quarter, data center revenue was $18.4 billion, more than five times that of the same period last year, accounting for 83% of Nvidia's total revenue.
The revenue of the automotive business was 281 million US dollars, a decrease of 4%, which is still a long way from the revenue of other core businesses and Huang Renxun's goals.
In order to sustain the growth of the automotive business, despite facing strict regulations from the United States on the export of advanced semiconductors to China, China, which has numerous emerging electric vehicle brands, remains a key market for Nvidia.
Nvidia has made many efforts in expanding into the Chinese market recently. In November last year, Nvidia issued a recruitment announcement on its official account, intending to expand its autonomous driving business in China. This is Nvidia's first large-scale recruitment for autonomous driving positions in China.
Nvidia's increased development of the Chinese autonomous driving market may yield significant results in the coming fiscal quarters.
Previously, Geely's Jike brand announced that it would use NVIDIA's DRIVE Thor technology. Nvidia has continued to expand its circle of friends with Chinese car companies at this GTC conference, and the competition in AI among car companies will become more intense in the future.