Bank of America Highlights Vera CPU as Nvidia's Next Major Growth Driver

Key Takeaways

• Bank of America estimates Nvidia's Vera CPU could generate about $20 billion in revenue during the second half of fiscal 2027.

• The opportunity is tied to growing demand for agentic AI and reinforcement learning workloads.

• Analysts expect roughly half of Vera CPU demand to support GPU systems, with the remainder coming from standalone CPU platforms.

• Nvidia could ship an estimated 4 million to 5 million Vera CPUs shortly after launch, according to Bank of America.

• The report suggests Nvidia's AI growth is expanding beyond GPUs into broader AI infrastructure.

What Happened

Bank of America has identified Nvidia Corp.'s (NASDAQ: NVDA) upcoming Vera CPU platform as a potentially significant new growth driver, arguing that the company's AI opportunity is expanding beyond graphics processing units (GPUs).

In a semiconductor industry note led by analyst Vivek Arya, the bank estimated that Vera CPU products could generate approximately $20 billion in revenue during the second half of Nvidia's fiscal 2027.

The analysis reflects expectations that enterprise AI workloads are evolving from model training toward large-scale deployment of autonomous or "agentic" AI systems, increasing demand for high-performance CPUs alongside GPUs.

Why It Matters

Nvidia has built its AI leadership primarily through GPUs used to train large language models and other artificial intelligence applications.

As AI adoption matures, however, computing requirements are becoming more diverse. Agentic AI systems, inference workloads, and reinforcement learning applications require close integration between CPUs and GPUs, creating an opportunity for Nvidia to expand its addressable market.

If Vera gains broad customer adoption, it could deepen Nvidia's position as a provider of complete AI computing platforms rather than individual processors.

Key Details

Bank of America's projections include:

MetricEstimatePotential Vera CPU revenueApproximately $20 billionExpected shipment volume4–5 million CPUsEstimated average selling price$4,000–$5,000 per CPU

According to the report:

• Approximately 50% of Vera CPU demand could come from processors paired with Nvidia GPU systems.

• The remaining demand may come from standalone CPU racks designed for agentic AI and reinforcement learning applications.

The analysis follows Nvidia's most recent quarterly results, during which Data Center compute revenue reached $60.4 billion, accounting for roughly 74% of the company's $81.6 billion in quarterly revenue.

Market Impact

The report reinforces the view that Nvidia's long-term growth may extend beyond its dominant GPU franchise.

Investors have largely focused on demand for Blackwell AI accelerators and hyperscale cloud spending. Bank of America's analysis suggests CPUs could become an additional multibillion-dollar business as enterprises deploy increasingly sophisticated AI workloads.

Future investor attention is likely to center on:

• Vera CPU launch timing.

• Enterprise adoption of agentic AI.

• CPU-GPU platform integration.

• Data center infrastructure spending.

• AI inference demand.

The estimates cited in the report represent analyst projections rather than company guidance, and actual results will depend on product adoption, customer demand, competitive dynamics, and broader AI infrastructure investment.

Conclusion

Bank of America believes Nvidia's Vera CPU platform could represent the company's next major AI growth engine, with the potential to generate approximately $20 billion in revenue as enterprise AI workloads evolve beyond model training. While GPUs remain central to Nvidia's business, expanding into high-performance CPUs could further strengthen its position as a full-stack AI infrastructure provider if customer adoption develops as anticipated.