Investing.com -- NVIDIA (NASDAQ: NVDA) has begun actively marketing its new Vera CPU to Chinese cloud and data-center clients, telling them the chip could be available as soon as August and that orders can already be placed, Reuters reported Friday, citing three sources familiar with the matter.

The sales push marks a strategic pivot for Nvidia as US export controls have effectively shut its advanced GPU lineup out of the Chinese market. CEO Jensen Huang has been candid about the damage: "Nvidia's market share in China has effectively fallen to zero," he said, a striking admission that underscores how urgently the company needs an alternative entry point. CPUs, which face fewer restrictions than Nvidia's high-end GPU products, offer exactly that — a largely unobstructed channel back into one of the world's largest data-center markets.

The early commercial interest is already tangible. According to Reuters, one major Chinese cloud company plans to order more than 300 servers, each containing two Vera CPUs, for initial testing before committing to larger-scale deployments. The pricing suggests significant revenue potential: a single Vera processor will cost "well north" of $20,000 before bulk discounts, and a fully configured rack of 256 chips would run approximately $10 million. Among the hyperscalers Nvidia is pitching, Crypto Briefing reported, are Alibaba and ByteDance, though neither company responded to requests for comment.

Huang has been laying the strategic groundwork publicly. During Nvidia's May 2026 earnings call he pegged the total CPU market opportunity at $200 billion and explicitly included China in that figure — a signal that the Vera push into the country is not opportunistic improvisation but a planned revenue lever. Deutsche Bank analysts, reviewing Huang's Computex 2026 keynote and analyst Q&A earlier this month, noted that the company is pursuing "a holistic approach to AI, with the newest strategic target area being CPUs to further the company's leadership in GPUs."

The market backdrop supports the ambition. Bank of America on June 11 raised its 2030 server CPU total addressable market estimate to more than $170 billion, up from $125 billion, implying a 37% compound annual growth rate over 2025-2030. Analyst Vivek Arya attributed the revision to agentic AI, calling it "a powerful demand accelerant that expands the CPU opportunity and lifts both x86 incumbents and ARM challengers." For Nvidia, which is entering a market long dominated by Intel and AMD, that expanding addressable market softens the competitive challenge — there may be enough growth for multiple winners.

The Vera chip itself is an Arm-based, 88-core processor built specifically for agentic AI workloads, unveiled in March 2026 and now in full production. It is part of Nvidia's broader Vera Rubin platform, which pairs the CPU with Rubin GPUs. The China pitch focuses on the CPU in isolation, precisely because the GPU component would likely draw regulatory scrutiny.

That regulatory question is one the industry is watching closely. US export controls have already delivered one false start in China for GPUs: Washington cleared H200 chip sales to ten Chinese firms, but those shipments stalled for months, according to reporting by Investing.com — a cautionary precedent that the Vera CPU's current lighter-touch classification could change. Whether the Bureau of Industry and Security will revisit that classification as Vera gains commercial traction in Chinese data centers remains an open and consequential question.

Software compatibility is another variable. Reuters sources noted that ecosystem and compatibility hurdles could slow large-scale adoption of the Arm-based chip in data centers historically built around x86 architecture — a friction point that the 300-server pilot orders appear designed to stress-test before buyers make deeper commitments.

Two calendar dates loom for investors seeking clarity. Nvidia's annual stockholder meeting on June 24 is likely to draw direct questions on China strategy, export-control exposure, and Vera's commercial ramp. That same day, Qualcomm is expected to make AI CPU announcements at its AI Day event, a reminder that the agentic-AI CPU race is still wide open and competitive pressure on Nvidia from the right flank is building even as it tries to reclaim ground in China.