NVIDIA's Spectrum-X Adoption Rises: Will It Boost Networking Revenues?

NVIDIA Corporation NVDA is seeing rising adoption of its Spectrum-X platform as artificial intelligence (AI) data centers grow larger and more complex. Spectrum-X is designed to enhance Ethernet networking performance for AI workloads, enabling graphics processing units (GPUs) exchange data faster and more efficiently. As more customers build large AI clusters, networking has become a critical part of overall system performance, making Spectrum-X an important growth lever for NVIDIA’s networking revenues.

In the third quarter of fiscal 2026, networking revenues surged 162% year over year to $8.19 billion, driven by strong demand for AI data center connectivity. As workloads become larger and more complex, the need for faster, more efficient interconnects is growing, and Spectrum-X is designed to meet that demand.

Cloud service providers and enterprise customers are increasingly deploying Spectrum-X to support training and inference tasks that involve thousands of GPUs. Companies like Meta Platforms, Oracle and CoreWeave are already integrating Spectrum-X to power their next-generation AI data centers. Additionally, other companies like Cisco and Hewlett Packard Enterprise are working with NVIDIA to integrate and offer the Spectrum-X architecture within their product lines and reference architectures for enterprise and cloud customers.

NVIDIA’s ability to integrate networking tightly with its GPU and software platforms gives it a unique advantage. Last year, the company introduced Spectrum-XGS, a next-generation version aimed at connecting multiple AI factories, showing its ambition to dominate large-scale networking for AI. If Spectrum-X adoption accelerates, networking could become one of NVDA’s most powerful revenue drivers in the AI infrastructure boom.

The Zacks Consensus Estimate for NVIDIA’s networking revenues in fiscal 2026 is pegged at $29.34 billion, indicating 125% growth from fiscal 2025’s $12.99 billion.

How Rivals Fare Against NVIDIA in AI Networking Space

NVIDIA faces competition from Broadcom AVGO and Arista Networks ANET in the AI networking space.

Broadcom is a leader in Ethernet switching and custom silicon solutions, providing high-performance chips that form the backbone of modern data centers. Broadcom’s strong relationships with hyperscalers like Google and Amazon allow it to deliver networking hardware tailored for large-scale AI and cloud workloads.

Arista Networks is another important rival, specializing in high-speed Ethernet switches used in AI and cloud environments. Arista Networks’ software-driven architecture and scalable systems make it a preferred choice for cloud service providers building AI clusters.

NVIDIA’s Price Performance, Valuation and Estimates

Shares of NVIDIA have risen around 21.1% over the past year compared with the Zacks Semiconductor – General industry’s gain of 25.9%.

NVIDIA One-Year Price Return Performance

From a valuation standpoint, NVDA trades at a forward price-to-earnings ratio of 24.58, lower than the industry’s average of 28.24.

NVIDIA Forward 12-Month P/E Ratio

The Zacks Consensus Estimate for NVIDIA’s fiscal 2026 and 2027 earnings implies a year-over-year increase of approximately 55.9% and 57%, respectively. Estimates for fiscal 2026 have been revised upward by 3 cents to $4.66 per share in the past 60 days. Earnings estimates for fiscal 2027 have been revised upward by 2 cents to $7.32 per share in the past seven days.

NVIDIA currently carries a Zacks Rank #1 (Strong Buy). You can see the complete list of today’s Zacks #1 Rank stocks here.

Want the latest recommendations from Zacks Investment Research? Today, you can download 7 Best Stocks for the Next 30 Days. Click to get this free report

This article originally published on Zacks Investment Research (zacks.com).

Zacks Investment Research