Nvidia is reportedly moving quickly to revive its position in China’s artificial intelligence market by promoting its new Vera central processing unit to major Chinese technology companies.

According to sources familiar with the discussions, Nvidia has informed prospective customers that orders for the Vera processor can now be placed, with shipments potentially beginning as early as August.

The outreach comes as Nvidia continues to face significant challenges in China following years of tightening U.S. export controls on advanced AI hardware.

Those restrictions have sharply reduced Nvidia’s ability to sell its most powerful graphics processing units into one of the world’s largest technology markets.

From GPUs to CPUs

For years, Nvidia’s dominance in artificial intelligence has been built primarily on graphics processing units, which remain the preferred hardware for training large AI models.

However, the company is now aggressively expanding into central processing units, particularly for AI inference workloads and autonomous AI agents.

Vera is Nvidia’s first standalone CPU specifically designed for agentic artificial intelligence systems, which can perform tasks and make decisions with minimal human intervention.

The company claims Vera can deliver performance levels up to 1.8 times faster than comparable competing processors in certain AI-related workloads.

This represents a major strategic expansion beyond Nvidia’s traditional GPU business.

China Shows Early Interest

Several Chinese cloud computing providers have reportedly expressed interest in testing Vera-based systems.

One large cloud operator is said to be evaluating an initial deployment of more than 300 servers equipped with dual Vera processors before deciding whether to pursue larger-scale adoption.

Industry observers note that early testing does not necessarily guarantee widespread deployment.

Software compatibility, infrastructure migration costs, and competition from domestic Chinese chipmakers remain important considerations for potential customers.

Nevertheless, the reported interest suggests Chinese firms remain eager to access advanced AI infrastructure where regulations permit.

A Different Regulatory Landscape

Unlike Nvidia’s high-end AI graphics processors, CPUs currently face fewer export restrictions.

This creates a potentially important opportunity for Nvidia to maintain a presence in China despite ongoing geopolitical tensions.

The company’s H200 AI accelerator, once expected to generate substantial sales in China, has faced prolonged shipment delays despite some licenses being granted.

Chinese authorities have also been encouraging greater adoption of domestic semiconductor technologies, further complicating Nvidia’s position.

By focusing on CPUs rather than GPUs, Nvidia may be attempting to navigate a more manageable regulatory environment while continuing to serve Chinese customers.

Challenging Intel and AMD

The Vera launch also intensifies competition with two of the industry’s largest processor manufacturers: Intel and Advanced Micro Devices.

Both companies have traditionally dominated the server CPU market through their x86 architectures.

Built on Arm-based technology, Vera represents Nvidia’s effort to establish itself as a significant competitor in the data center processor segment.

The timing may be advantageous.

Industry reports indicate that global demand for AI-focused server processors continues to exceed supply, creating shortages and extended delivery times across portions of the market.

As AI infrastructure spending accelerates worldwide, alternative processor providers could find substantial opportunities for growth.

AI Infrastructure Spending Continues to Expand

The launch comes as hyperscalers, cloud providers, and enterprise customers continue investing aggressively in AI infrastructure.

The industry’s focus is increasingly shifting beyond model training and toward inference computing, where AI systems generate responses, execute tasks, and support autonomous applications.

This transition is creating growing demand for CPUs, custom silicon, networking hardware, and supporting infrastructure alongside traditional GPUs.

Nvidia believes Vera can play a central role in this next phase of AI deployment.

The company reportedly expects the product line to generate approximately $20 billion in revenue by the end of its current fiscal year.

Cost and Deployment Considerations

The Vera platform is expected to target large-scale enterprise and cloud deployments.

Industry estimates suggest individual processors may cost well above $20,000 before volume discounts, while fully configured AI racks could reach several million dollars depending on system specifications.

Such pricing positions Vera as a premium infrastructure solution aimed primarily at hyperscale operators and major enterprise customers.

For Chinese clients, initial deployments are reportedly expected to occur primarily in overseas data centers while performance and compatibility are evaluated.

The Bigger Picture

Nvidia’s Vera initiative highlights how quickly the company is adapting to changing market conditions.

Faced with restrictions on GPU exports and increasing competition from domestic Chinese semiconductor firms, Nvidia is broadening its product strategy while pursuing new opportunities within AI infrastructure.

Whether Vera achieves large-scale adoption remains uncertain, but the product demonstrates Nvidia’s determination to remain a major force in global AI computing even as regulatory and competitive pressures continue to evolve.

For investors and industry observers, Vera represents more than a new processor. It signals Nvidia’s effort to build multiple growth engines beyond GPUs as the next phase of artificial intelligence infrastructure development unfolds.