Nvidia, the world's No. 1 companies by market capitalization, set all-time highs for both quarterly revenue and operating profit on the back of growing demand for artificial intelligence (AI) semiconductors. Revenue set a new record for the 13th straight quarter.
Nvidia said on the 26th (local time) that revenue in the second quarter (May–July) of fiscal year 2027 was $96.221 billion (about 133.36 trillion won) and operating profit was $63.734 billion (about 88.34 trillion won). Those were up 106% and 124%, respectively, from a year earlier. Compared with the prior quarter, revenue rose 18% and operating profit increased 19%.
Revenue topped the market consensus of $92.17 billion (about 127.75 trillion won) compiled by London Stock Exchange Group (LSEG) by more than $4 billion. On a non-GAAP adjusted basis, diluted earnings per share (EPS) were $2.22 (about 3,077 won), above the market estimate of $2.10 (about 2,911 won).
Net income under U.S. Generally Accepted Accounting Principles (GAAP) was $59.688 billion (about 82.73 trillion won), up 126% from a year earlier. Diluted EPS was $2.46 (about 3,410 won). The gross margin was 75.0%. The operating margin, the ratio of operating profit to revenue, was about 66.2%.
Growth was driven by the data center business, which includes graphics processing units (GPUs) for AI servers and networking equipment. Data center revenue was $89 billion (about 123.35 trillion won), up 18% from the prior quarter and 117% from a year earlier. That accounts for about 92.5% of total revenue. Edge computing revenue was $7.2 billion (about 9.98 trillion won), up 13% from the prior quarter and 27% from a year earlier.
Nvidia guided third-quarter (Aug.–Oct.) fiscal year 2027 revenue of $105.84 billion to $110.16 billion (about 146.69 trillion to 152.68 trillion won). That exceeds the $104.19 billion (about 144.41 trillion won) market consensus compiled by LSEG. Nvidia said this outlook does not include revenue from China data center computing.
However, it expects profitability to edge down. It guided both GAAP and non-GAAP gross margins for the third quarter at 73.5%–74.5%. That is lower than the second quarter's 75.0% and below the 74.77% market consensus compiled by LSEG.
Jensen Huang, Nvidia chief executive officer (CEO), said, "AI has reached an inflection point," and added, "Computing is now generating revenue and demand is accelerating." He noted, "AI infrastructure buildout is proceeding at full speed," and said the next-generation AI platform "Vera Rubin" has entered full mass production.
Nvidia said racks based on Vera Rubin are running at CoreWeave, Google Cloud, Microsoft Azure, Oracle Cloud Infrastructure (OCI), and Nevius.
Beyond selling AI Semiconductor, Nvidia is also involved in financing for data center buildouts. It is partnering with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs, and KKR to raise more than $500 billion (about 693 trillion won) in external capital for AI infrastructure. The plan assumes final contract execution.
In Korea, it is pursuing gigawatt (GW)-scale AI infrastructure with SK Telecom, Naver, and Brookfield. The plan is to build a "sovereign AI" infrastructure that allows a state to control its own data and computing base on the Nvidia DSX platform. With SK hynix, it has formed a multiyear collaboration to develop next-generation memory technologies for AI data centers worldwide.