Nvidia on Thursday denied a report by The Information that it planned to begin shipping a language processing unit (LPU) designed for Chinese customers by year-end. An Nvidia spokesperson said the company has no China-specific LPU product on its roadmap and no LPU sales in the Chinese market at present.
The denial follows a March report by Reuters that Nvidia was developing a China-compatible version of its AI chips for the inference market. The company has also faced restrictions in China after Washington approved limited sales of its H200 AI chips to a small group of Chinese firms in May, including Alibaba, Tencent, and ByteDance.
Nvidia Chief Executive Jensen Huang acknowledged in May that the company had "largely conceded" the Chinese AI chip market to domestic rival Huawei Technologies. The LPU in question was developed using technology licensed from startup Groq and is designed to work alongside GPUs to accelerate AI chatbot responses.
The Information had claimed the reported China-specific LPU version complied with U.S. export-control rules, featuring modified software to operate with processors available in China following the unavailability of Nvidia's next-generation Vera Rubin AI system due to U.S. restrictions.
