Nvidia has begun pitching its new Vera central processing unit (Vera AI CPU ) to Chinese customers, signaling a strategic effort to regain momentum in a market where its position has been severely weakened by U.S. export restrictions. According to three sources familiar with the matter, the company has informed potential clients in China that the Vera processor, designed for artificial intelligence (AI) data centres, could be available as early as August and that customers can now begin placing orders.
The move highlights Nvidia’s determination to revive its business in China after shipments of its H200 AI CPU chip—the company’s second-most powerful AI processor—have remained stalled for months. The situation has significantly impacted Nvidia’s presence in the country, with CEO Jensen Huang stating in October that the company’s market share in China had effectively fallen to zero. The decline has been driven by tightening U.S. export controls on advanced semiconductor technology as well as Beijing’s growing push for technological self-sufficiency.
Vera AI CPU represents a major step in Nvidia’s AI strategy. It is the company’s first standalone CPU built specifically for the emerging field of agentic AI, where systems are designed to perform tasks autonomously with limited human intervention. Sources said several Chinese customers have already expressed interest in the new processor, although the discussions remain private.
Now in full production, Vera is intended to handle the core computing tasks that support Vera AI CPU agents behind the scenes. Nvidia claims the chip can deliver up to 1.8 times the performance of comparable processors offered by competing manufacturers. The launch also places Nvidia in more direct competition with traditional CPU leaders Intel and AMD, both of which are racing to expand their presence in AI-focused data centres as demand for computing power continues to surge worldwide.
Nvidia first unveiled Vera in March, when Huang described the processor as a product with the potential to become the company’s next multibillion-dollar business. At the time, Nvidia announced that major cloud and technology companies, including Alibaba and ByteDance, were collaborating with the company to deploy the new chip. However, Nvidia did not indicate whether formal orders had already begun.
The company declined to comment on the latest developments. Alibaba and ByteDance also did not respond to requests for comment. If successful, Vera could become a key part of Nvidia’s efforts to rebuild its presence in China while strengthening its position in the increasingly competitive AI infrastructure market.
Also Read:- Why CUDA is Key to the Future of Autonomous Vehicles in America?
