Recently, He Xiaopeng, chairman of XPeng Motors, revealed in an interview that the company’s self-developed Turing AI chip will officially enter mass production in the second quarter of 2025. This chip will also be applied to XPeng’s AI robots and flying cars, enabling physical AI hardware.
XPeng Motors announced the successful tape-out of the self-developed Turing chip last August, featuring 40 cores capable of running a 30B parameter AI large model, designed for L4 autonomous driving. It can be applied simultaneously in AI robots, AI cars, and flying cars, claiming to outperform NVIDIA’s Orin X by a factor of three, capable of driving both autonomous driving and intelligent cockpit large models. By October of the same year, XPeng had already implemented intelligent driving functions on the Turing chip.
XPeng Motors stated that through deep customization, the Turing chip’s computing power can achieve 100% optimal utilization, and future AI cars will be equipped with at least three Turing chips.
According to market news, XPeng Motors will launch a brand new model at the end of May or early June this year, which will be the first model equipped with the self-developed chip.
Insiders claim that the computing power of a single Turing chip is significantly stronger than NVIDIA’s Orin X at 254 TOPS, only slightly lower than the mid-range version of NVIDIA’s Thor, which has not yet entered mass production. “(The computing power of the Turing chip) is sufficient for XPeng, and its cost-performance ratio is much higher than that of Thor,” it was revealed. It is reported that all new models from XPeng Motors will be equipped with the self-developed chip and will no longer plan to use Thor chips.
END.
Previous Recommendations
Global smartphone shipments slightly increased in Q1, as manufacturers accelerate diversification under tariff shadows
Foxconn reportedly intends to establish the world’s second-largest production base in India
Q1 sales ranking of new force models: Xiaomi SU7 leads
Huawei issues bonds three times in a month, raising a total of 12 billion yuan
Lotus Cars announces layoffs of up to 270 people in the UK
NVIDIA announces the establishment of a domestic AI server supply chain in the US, aiming for a production value of $500 billion in four years
*Disclaimer: This article is original by the author, and the reprint by 52RD is to share this information or viewpoint, and does not represent support for the viewpoint. If there are any objections, please feel free to contact us.