Nvidia attracted attention with the strong financial results it announced in the third quarter of the 2025 fiscal year. The company generated $57 billion in revenue during this period. This figure indicates a significant increase of 62 percent compared to the same period of the previous year. According to GAAP standards, net income was announced as 32 billion dollars; This corresponds to a 65 percent increase on an annual basis. Exceeding analyst estimates in both revenue and profitability, Nvidia consolidated its leading position in the global artificial intelligence market.
Behind this impressive performance of the company is the strong contribution of its data center business unit. Nvidia’s data center unit broke its own record with $51.2 billion in revenue this quarter. This figure means an increase of 25 percent compared to the previous quarter and 66 percent compared to the same period last year. The remaining share of total revenue came from gaming hardware sales of $4.2 billion, and the rest came from professional visualization and automotive segments.
Nvidia drew attention to the growth in data centers
Colette Kress, the company’s CFO, explains this growth in the data center segment; accelerating computing needs, advanced artificial intelligence models and new generation applications. Speaking at the third quarter investor information meeting, Kress stated that artificial intelligence infrastructure and factory projects corresponding to a total of 5 million GPUs were announced during this period. In this context, it was emphasized that the demand from cloud service providers (CSP), public institutions, large corporate customers and supercomputer centers remains at a high level.
The Blackwell Ultra GPU series, which Nvidia announced in March, has risen to the strongest position in the company’s internal structure. While demand for these next-generation chips far exceeds the company’s expectations, previous versions of the Blackwell architecture continue to attract high interest. CEO Jensen Huang stated that Blackwell sales have reached extraordinary levels and that cloud-based GPUs are completely sold out. “The demand for computing for training and inference is growing exponentially. The AI ecosystem is rapidly scaling with more sectors, countries and initiatives,” Huang said.
However, the company’s situation in the Chinese market fell short of expectations. Shipments of the data center GPU called H20, developed for generative artificial intelligence and high-performance computing, remained at 50 million units. Kress explained that the main reasons why this figure remains low are geopolitical restrictions and increased competition in the Chinese market. “We continue to engage with the U.S. and Chinese governments and will continue to strive to keep American companies globally competitive,” Kress said.
In light of all these developments, Nvidia’s predictions for the fourth quarter are also quite positive. The company expects revenue of $65 billion in the last quarter of the year. This strong prediction was also reflected in share prices; Nvidia shares rose more than 4 percent in after-market trading.
In his statement to investors, Huang emphasized that the growth in the field of artificial intelligence is not a temporary fad. “There is talk of an artificial intelligence bubble, but we see a completely different picture,” he said. The company’s statements and figures show that a much deeper transformation process is in progress than short-term speculations.