Nvidia Posts $115.8 Billion Quarter, Raises Outlook as Huang Exits

Nvidia booked $115.8 billion in quarterly revenue, beating its own $108 billion guidance, even as its stock has fallen as much as 39% in a little more than a month. On Tuesday, CFO and interim CEO Collette Kress and new Chairman of the Board Harvey Jones hosted the Q3 earnings call. It was the first such call without founder Jensen Huang, who resigned as CEO last week.
The numbers remain absurd by any historical standard. Nvidia guided to $130 billion for next quarter, a period that includes one extra week, which puts the company on a pace of nearly half a trillion dollars a year. Management also raised its growth outlook for next year to about 75 percent, up from 70 percent. That is a guidance raise during a month in which the market shaved more than a third off the share price. The arithmetic and the sentiment have not yet agreed on a story.
What changed with Huang's departure?
For three years, Jensen Huang decided who received the scarcest chips in the world. Now, as macro investor Bryan Rich of Logic Fund Management put it, it sounds like the capital markets will decide — and the capital markets take their cue from the Fed. The board has confirmed it will not pursue Huang's return as CEO, closing the door on the founder who presided over the entire AI buildout.
The spenders themselves show no sign of slowing. The four biggest — Microsoft, Google, Amazon, and Nvidia — are on pace for roughly $730 billion of spending this year. On Tuesday, Nvidia said the cloud industry's backlog of signed business now stands at $2.5 trillion, up from $2 trillion just three months ago. A $500 billion addition to committed demand in one quarter is not a cooling market. It is a market that has simply outrun the patience of equity investors.
Are the chips still scarce?
Yes, and the pricing evidence is blunt. Token prices have fallen nearly 60 percent since May even as usage has surged — a deflationary signal for consumers of AI, not for sellers of silicon. Nvidia says its chips are fully utilized at its major cloud customers, and older-generation chips have been re-contracted at up to twice their original price. Hardware from two cycles ago now commands a premium over its launch price. Scarcity does not usually announce itself more clearly than that.
Where is the risk now?
The risk has migrated from demand to balance sheets. Nvidia has extended guarantees covering up to $118 billion of partner obligations, some of them contingent on financing and capacity actually coming online. That figure deserves attention: it approaches an entire quarter of revenue committed as backstop, and it makes Nvidia a quiet lender to its own ecosystem.
Management is responding by scrutinizing future commitments and shifting its capital focus toward expanding supplier capacity. The new terms are simple. Customers must have power and capital ready to secure chips. Demand alone no longer wins an allocation; the balance sheet does. In that sense, Huang's exit and the Fed's fingerprints on chip allocation are the same story told twice.
The company that beat guidance by $7.8 billion and raised its growth outlook by five points will now be measured less on hyperscaler appetite, which shows $2.5 trillion of signed backlog, and more on whether financed capacity and guaranteed obligations convert into collected revenue.
Source: seekingalpha.com
Nathan Brooks
Show full bio
Senior reporter covering people, launches and day-to-day practice in Semiconductors, chip manufacturing, hardware, AI compute at Chip Dispatch.




