[News] NVIDIA Earnings Spotlight: $20B CPU Sales Target in 2026, China Concession to Huawei, and Reporting Overhaul

While NVIDIA posted another record quarter for the three months ended late April with revenue of $81.6 billion, the company forecast further growth to $91 billion for the current quarter, above analyst expectations.

However, beyond the headline numbers, the earnings call has pointed to a broader strategic shift, including positioning its Vera CPU as a key new growth driver, acknowledging Huawei’s growing role in China’s AI chip market, and restructuring its revenue reporting into a new platform-based framework separating data center and edge computing businesses.

NVIDIA Expands Into CPU Market With Vera Push

As Intel and AMD benefit from a surge in CPU demand driven by the rise of agentic AI, CNBC, citing CFO Colette Kress, reports that NVIDIA is aiming to become the “world’s leading CPU supplier.”

Kress said the company’s new Vera CPU “opens a brand new $200 billion opportunity,” adding that “every major hyperscaler and system maker is partnering with us to get it deployed.” According to Kress, NVIDIA’s CPU revenue could reach $20 billion in 2026, CNBC reports.

The strategy marks a clear expansion beyond GPUs, building on remarks from NVIDIA’s GTC conference in March, where VP of hyperscale and HPC Ian Buck said the company would focus on a single Vera CPU SKU designed specifically for agentic AI workloads tightly integrated with its GPU platform, as reported by Tom’s Hardware.

Rather than competing head-on in the traditional CPU market, Vera is tuned for AI-centric compute, prioritizing single-thread performance over high core counts and taking a different architectural path from AMD’s EPYC and Intel’s Xeon lines, according to the report.

Notably, Ian Buck personally delivered the first batch of Vera CPUs to key cloud and AI partners, including Anthropic, OpenAI, SpaceX, and Oracle Cloud, the company said in an X post on Monday, signaling early adoption from major hyperscaler customers.

NVIDIA described Vera as its first custom CPU purpose-built for the era of agentic AI, underscoring its push to anchor the chip within large-scale cloud deployments rather than traditional CPU markets, as highlighted by PC Mag.

No Hopper Shipments as Huawei Gains Ground in China AI Market

On the other hand, NVIDIA seems to be much more conservative for its China outlook. According to Yahoo! Finance, the company reported no China-related revenue during the quarter, while confirmed it did not ship any Hopper products to the market, compared with $4.6 billion in China revenue in the same period last year.

CEO Jensen Huang told CNBC that demand in China remains “quite large,” noting that Huawei has grown “very, very strong” and is likely to deliver another “extraordinary year,” supported by a rapidly expanding domestic chip ecosystem.

However, He also said NVIDIA has “really largely conceded that market to them.” According to CNBC, China once accounted for at least one-fifth of NVIDIA’s data center revenue.

While the U.S. has approved export licenses for the tightly restricted H200 chips, Beijing has reportedly discouraged local firms from purchasing them as China accelerates efforts to strengthen its domestic semiconductor ecosystem.

NVIDIA Revenue Structure Redrawn

As highlighted by Yahoo! Finance, NVIDIA is also overhauling its reporting structure, reorganizing the business into two main platforms: Data Center and Edge Computing. According to the report, the Data Center segment will cover hyperscalers and dedicated AI data center operations, while Edge Computing will include agentic and physical AI-related devices such as AI PCs, gaming consoles, workstations, AI-RAN infrastructure, robotics and automotive platforms.

The company said the restructuring is intended to give investors a clearer view of its long-term growth engines. Previously, NVIDIA reported results across separate divisions including Data Center, Gaming and AI PC, Professional Visualization, and Automotive and Robotics, the report adds.

As noted by CNBC, hyperscalers accounted for more than half of NVIDIA’s total data center revenue, reaching $38 billion and growing 12% sequentially.

The remaining $37 billion came from a segment NVIDIA refers to as ACIE, covering AI cloud, industrial and enterprise-related businesses, the report notes, adding that AI cloud revenue more than tripled from a year earlier.

Read more

• [News] U.S. Reportedly Approves NVIDIA H200 Sales to 10 Chinese Firms Including Alibaba, ByteDance, but Deals Stall
• [News] NVIDIA Reportedly Plans GPU-Direct Storage for Vera Rubin, Raising Expectations for HBF Beyond HBM

(Photo credit: NVIDIA)