The numbers Nvidia posted Wednesday evening after the closing bell were not just a beat — they were a redefinition of what a technology company can generate in a single quarter. Record revenue of $81.6 billion, up 85% year over year. Data center revenue of $75.2 billion, up 92%. Net income of $58.3 billion — a 211% increase from a year ago. Non-GAAP earnings per share of $1.87, clearing the $1.77 consensus estimate. Gross margins held at 75% despite a simultaneous transition between two major chip architectures.
And then came the guidance. Nvidia is projecting $91 billion in revenue for the current quarter — well above the $87 billion Wall Street consensus and comfortably ahead of the highest whisper numbers circulating before the print. The company announced a new $80 billion share repurchase authorization and returned approximately $20 billion to shareholders through buybacks and dividends in the quarter alone.
Nvidia’s stock rose modestly after hours, a reflection not of disappointment but of a market that had already priced in excellence and received confirmation.
What’s Driving It
The engine behind the numbers is Blackwell — Nvidia’s current generation AI chip architecture that now drives the majority of data center compute revenue. Blackwell 300 products ramped aggressively in the quarter, and Nvidia’s networking solutions — including InfiniBand, Spectrum-X Ethernet, and NVLink — posted networking revenue growth of 64% sequentially as AI factories scaled their interconnect infrastructure.
Nvidia also launched the Vera Rubin platform during the quarter — its next-generation architecture purpose-built for agentic AI workloads. The Vera CPU is described as the world’s first processor designed specifically for AI agents, with first deployments expected at Amazon Web Services, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure, and CoreWeave in the second half of 2026. At its March GTC conference, CEO Jensen Huang projected that Blackwell and Vera Rubin combined would generate $1 trillion in revenue across 2026 and 2027. Wednesday’s results do nothing to undermine that projection.
Notably, Nvidia’s Q2 guidance explicitly excludes any data center compute revenue from China — the H20 export restrictions imposed in April remain fully in effect — making the $91 billion outlook that much more significant.
The Small and Microcap Read-Through
For investors operating below the $2 billion market cap threshold, Nvidia’s quarter is not just a large-cap story. It is a forward demand signal for an entire ecosystem of smaller companies.
The top five hyperscalers — Amazon, Microsoft, Google, Meta, and Oracle — are now expected to nearly double their capital expenditure spending in 2026, a significant revision upward from prior estimates of 62% year-over-year growth. That level of infrastructure commitment does not get executed through Nvidia alone. It flows through hundreds of suppliers, component manufacturers, and technology providers operating at every layer of the AI buildout stack.
Smaller companies in specialty semiconductor materials, advanced cooling systems, power infrastructure, optical networking components, and AI-optimized software are direct downstream beneficiaries of a sustained hyperscaler capex cycle. Many of those companies sit well below the $2 billion market cap threshold and have yet to see their valuations reflect the demand environment Nvidia’s results just confirmed.
The AI infrastructure buildout is not slowing. Wednesday night’s print made that case with $81.6 billion worth of evidence.
