Nvidia rejects claims of China-specific AI chip launch by year-end

Nvidia has pushed back against reports that it is preparing a new AI processor specifically for the Chinese market, saying no such product is currently part of its roadmap.

“The reporting in The Information on Nvidia’s LPU is incorrect. We have no LPU sales in the China market today, and no China-specific LPU product in our roadmap,” an Nvidia spokesperson said.

The denial followed a report claiming that it planned to begin small-batch shipments of a language processing unit (LPU) designed for Chinese customers by the end of the year. The report, citing 2 Nvidia employees, said several customers had already placed orders.

The reported chip was described as an LPU based on technology licensed from startup Groq. It was said to work alongside graphics processing units (GPUs) to speed up AI chatbot responses.

The report also claimed that the processor would comply with US export-control rules. Nvidia was reportedly modifying its software so the LPU could work with processors available in China after its next-generation Vera Rubin AI system became unavailable in the market due to US restrictions.

Earlier reports had indicated that Nvidia was preparing a China-compatible AI chip as it looked to compete in the growing AI inference market.

Nvidia’s China strategy remains in flux

Nvidia’s business in China has faced major changes in recent months. In May, Washington approved sales of the company’s H200 AI chips to a limited group of Chinese companies, including Alibaba, Tencent and ByteDance. Deliveries of those chips have only recently begun.

Nvidia has also been promoting its Vera CPU to Chinese customers.

CEO Jensen Huang said in May that Nvidia had “largely conceded” China’s AI chip market to Huawei Technologies, which has emerged as its main domestic competitor.