Nvidia outpaces AMD as AI chip demand drives 106% revenue surge

Nvidia reported $96.2 billion in fiscal Q2 revenue, up 106% year over year, cementing its lead over AMD in the AI chip race.

"Demand is much greater than 70%," CEO Jensen Huang said, referring to fiscal 2028 growth projections that CFO Colette Kress set at approximately 70%.

AMD's Q2 revenue reached $11.5 billion, up 50% year over year, with data center sales more than doubling. Nvidia's data center segment generated $89 billion, up 117% year over year, representing roughly 92% of total revenue.

Nvidia trades at 23.5x forward earnings versus AMD's 64.2x, with ROE of 99.7% compared to AMD's 14.9%. Nvidia guides Q3 revenue to $108 billion, while AMD expects $13 billion.

The divergence in stock performance began in early July, when Nvidia, the world's largest company by market value, started outpacing AMD on the charts. The gap reflects more than just revenue scale — it captures the structural advantages Nvidia holds in software, pricing power, and customer lock-in.

Nvidia's CUDA software ecosystem, refined over more than a decade, creates switching costs that AMD's ROCm platform has yet to match. Cloud providers including Microsoft, Amazon, and Google continue to deploy Nvidia GPUs at massive scale, with Meta alone narrowing its 2026 capital expenditure range to $130-145 billion, much of it directed at AI infrastructure featuring Nvidia silicon.

Nvidia's Vera Rubin platform is in full production, with systems sold out through mid-year and a backlog extending into fiscal 2028. The company's gross margins held at 75% on both GAAP and non-GAAP bases, up from roughly 72.5% a year ago, demonstrating sustained pricing power in a supply-constrained market.

AMD is making progress with its EPYC server processors and Instinct accelerators, and the Helios platform has begun initial ramp-up. The company projects a non-GAAP gross margin of 56% in Q3, reflecting ongoing margin expansion. But AMD's return on equity of 14.9% trails Nvidia's 99.7% by a wide margin, and its forward P/E of 64.2x implies investors are paying a substantial premium for a company still fighting for market share.

The competitive threat extends beyond AMD. Google has invested heavily in its TPU program, Amazon continues to develop Trainium and Inferentia chips, and Microsoft has reportedly begun work on custom AI accelerators. These initiatives reflect hyperscalers' desire to reduce dependence on Nvidia, though the CUDA ecosystem's network effects have so far kept most enterprise workloads on Nvidia hardware.

Nvidia's guidance excludes any China-related sales, meaning potential sales from the region could provide additional upside. The company's Zacks Rank of #2 (Buy) versus AMD's #3 (Hold) reflects the relative strength of the two investment cases.

For investors, the choice between NVDA and AMD comes down to whether they want the proven leader or the challenger. Nvidia's combination of 106% revenue growth, 75% margins, and a 23.5x forward P/E offers growth at a reasonable valuation. AMD's 50% growth and improving margins are respectable, but the 64.2x multiple prices in a level of success the company has yet to demonstrate. Investors will watch Nvidia's Q3 earnings report, expected in November, to test whether the $108 billion guidance holds.

This article is for informational purposes only and does not constitute investment advice.