Following successful efforts to secure approval from the Trump administration for selling its H200 chips in China, Nvidia is now evaluating an increase in production as Chinese firms accelerate their order placements. According to a report citing unnamed sources, these H200 graphics processing units (GPUs)—the most capable chips from Nvidia’s earlier Hopper series designed for training large language models—were previously barred from the Chinese market. This restriction stemmed from proposed rules under the prior Biden administration aimed at limiting sales of advanced artificial intelligence chips to the country.

Last week, however, the Department of Commerce granted Nvidia permission to sell H200 GPUs in China, contingent on the company surrendering 25% of the sales revenue from those chips. Demand from Chinese companies is reportedly so robust that Nvidia is exploring options to expand its manufacturing capacity. Still, Chinese authorities have yet to finalize a decision on whether to permit imports of the H200 chips, which are understood to be substantially more powerful than the tailored H20 GPUs Nvidia had developed for the Chinese market.

For Nvidia, boosting production of the H200 GPUs would enable it to address pent-up demand in a nation aggressively pursuing its own domestic AI chip development. Western competition and national security considerations have constrained access to the newest and most potent hardware for AI training in China, leading companies there to prioritize efficiency over raw scale. Major Chinese tech firms like Alibaba and ByteDance, which are crafting their own AI models, have already contacted Nvidia to discuss substantial orders for the H200 chips, which are currently being manufactured in limited volumes.

In an emailed statement, an Nvidia spokesperson clarified, “We are managing our supply chain to ensure that licensed sales of the H200 to authorized customers in China will have no impact on our ability to supply customers in the United States.”