#我的七夕交易分享

NVIDIA’s next big test may have very little to do with whether it beats the next earnings estimate.

The market has already become accustomed to NVIDIA delivering numbers that look exceptional. The harder question now is whether AI infrastructure spending can keep expanding fast enough to support another major valuation cycle.

The latest numbers show why expectations are so high. NVIDIA reported $81.6 billion in fiscal Q1 2027 revenue, up 85% year over year, while Data Center revenue reached $75.2 billion, up 92% YoY. Gross margin remained around 75%, and management guided fiscal Q2 revenue to approximately $91 billion.

So another earnings beat alone may not be enough to surprise investors.

The bigger story is Vera Rubin.

Rubin is moving from a future product narrative toward real infrastructure deployment. NVIDIA says Vera Rubin systems are already ramping across partners including CoreWeave, Google Cloud, Microsoft, Oracle and Nebius, with the platform designed around significantly better performance-per-watt and lower inference costs.

That changes the investment debate.

AI demand is increasingly shifting from simply training larger models toward continuous inference, agentic AI and production workloads. If Rubin can lower the cost of generating AI tokens while increasing useful compute per unit of power, it could encourage customers to deploy even more AI capacity rather than simply replace existing hardware.

And there is another variable that deserves attention: who finances the next wave of AI factories?

NVIDIA recently announced partnerships involving Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR aimed at mobilizing more than $500 billion of third-party capital over time for AI infrastructure. That suggests the next phase of the AI boom may increasingly resemble infrastructure investment rather than conventional corporate IT spending.

But the bullish case is not risk-free.

Higher memory, wafer, packaging and networking costs can pressure margins. AMD and custom ASICs remain competitive threats, while hyperscalers continue developing their own silicon. If customers eventually decide that proprietary chips deliver better economics for specific workloads, NVIDIA’s market-share advantage could face pressure.

That makes margin quality and return on AI investment just as important as revenue growth.

The most interesting signal from NVIDIA’s latest results is therefore not simply the size of the revenue number. It is the combination of 85% YoY revenue growth + roughly 75% gross margins + a $91 billion quarterly outlook + Rubin entering real-world deployment.

My view: the next NVIDIA cycle will be judged on whether the company can turn extraordinary AI demand into lower-cost, higher-scale computing economics.

If Rubin succeeds, NVIDIA may be extending the AI semiconductor cycle rather than simply replacing Blackwell.

If margins weaken, custom chips gain traction and AI infrastructure returns fail to justify spending, the market could become much less willing to pay a premium for future growth.

The next earnings report will give us the numbers.
Rubin will give us the bigger answer. $NVDA
#UnitreeTechSoars629%OnDebuts