On Wednesday after post-market trading, the global company with the highest market cap, referred to by Goldman Sachs as "the most important stock on Earth," $NVIDIA (NVDA.US)$ released its earnings for the third fiscal quarter of fiscal year 2025. This leading AI chip company continues to experience high growth, with Q3 revenue growth rate slightly slowing but still recording 94%, and net income doubling.
However, due to nvidia's fourth-quarter revenue guidance mid-point growth of 70%, although exceeding the average Wall Street expectations, it did not reach the highest expected growth rate of 90%, causing a significant shake-up in post-market stock prices. As of the time of filing, nvidia reversed gains, now up over 1%.
The demand for AI shows no signs of decline, and the growth rate for datacenters continues to explode.
Overall, Nvidia's performance this time remains outstanding. Q3 revenue was $35.082 billion, a year-on-year increase of 94%; the adjusted EPS was $0.81, a year-on-year increase of 103%, both exceeding market expectations. Net income increased by 109% year-on-year to $19.3 billion; the adjusted gross margin for Q3 was 75%, consistent with the level a year ago and in line with analyst expectations, with a 0.7 percentage point decrease sequentially.
From a business perspective, the datacenter segment, which contributes the vast majority of revenue to Nvidia, continues to "explode" with segment revenue reaching an astonishing $30.8 billion, a year-on-year increase of 112%, compared to 154% growth in the previous quarter; over the past four years, Nvidia's datacenter business has grown approximately 27 times, with the total revenue contribution rising from 37% to 88%, making it a unique core segment.
Nvidia's CFO Kress commented that the year-on-year and quarter-on-quarter growth in the datacenter segment in Q3 was due to market demand for the Hopper computing platform, with cloud computing service providers contributing about 50% of Nvidia's datacenter revenue, while the remaining revenue came from consumer internet companies.
It can be seen that the global trend in laying out ai infrastructure has not waned, and major technology giants are still "spending money" on ai, supporting nvidia's strong profitability. Currently, the revenue of nvidia's datacenter department alone exceeds the combined revenue of its two major competitors—intel and amd.
Revenue from other departments also continues to maintain double-digit growth. Revenue from the gaming and ai pc business is 3.3 billion dollars, a year-on-year increase of 15%; professional visualization revenue is 0.486 billion dollars, a year-on-year increase of 17%; and revenue from the autos and siasun robot&automation business is 0.449 billion dollars, a year-on-year increase of 72%.
Despite the overall performance exceeding expectations, why did nvidia's stock price fall?
In the past five years, nvidia has only had one quarter where its performance did not meet the average analyst expectations. Moreover, in recent quarters, its actual total revenue has even exceeded Wall Street expectations by at least 20%.
It seems that the remarkable revenue growth rate of nvidia in the past two years has led analysts to set extremely "strict" performance standards for it. This quarter, nvidia's revenue growth rate reached 94%, but compared to the growth rates of 122%, 262%, and 265% in the previous three quarters, it is still a slowdown.
In addition, nvidia's guidance for the next quarter also failed to impress the market. Looking ahead to the fourth quarter of fiscal year 2025, the company expects revenue of 37.5 billion dollars, with an upward and downward fluctuation of 2%, slightly above the market average expectation of 37.08 billion dollars but below the highest expectation of 41 billion dollars. In terms of growth rate, nvidia's Q4 guidance indicates a year-on-year increase of 70%, but Wall Street's highest expectation is 90%.
Before nvidia released its financial report, Wall Street had already shown a divergence in views on nvidia's Blackwell product line. Dan Eye, Chief Investment Officer of Fort Pitt Capital Group, stated: "There are many unknowns regarding Blackwell's production capacity. (Nvidia's) CEO has already established a high level of credibility, but the threshold is very high." He expects that nvidia may find it difficult to provide astonishing guidance for the fourth quarter.
Production capacity ramp-up is going smoothly! Jensen Huang has dispelled the "Blackwell concerns".
Currently, the market's primary focus is on nvidia's Blackwell chip. In a conference call, CEO Jensen Huang stated that the design flaws of the Blackwell architecture AI GPU have been completely resolved, and it is now in full production. He emphasized that the demand for this high-performance AI GPU product is expected to exceed projections for several consecutive quarters. Huang also proclaimed in the earnings statement that the market demand for Blackwell and Hopper is 'incredible.'
The company's chief financial officer Colette Kress stated that nvidia successfully replaced the mask for Blackwell, improving production yield. The company plans to start shipping Blackwell chips in the fourth quarter of this fiscal year and will accelerate shipments throughout the entire fiscal year 2026. It is anticipated that there will be a shortage of Blackwell chips in several quarters of fiscal year 2026, the next financial year.
The latest statements from these executives undoubtedly injected confidence into the market, alleviating concerns over the delays in Blackwell due to design defects, which is also the main reason for the narrowing of the stock price decline in post-market trading.
Moreover, the company mentioned that the launch of Blackwell this quarter will lead to increased costs, which will reduce the gross margin. In the fourth quarter, demand for nvidia's Hopper will continue, and Blackwell will see initial growth, with revenue expected to exceed previously projected billions of dollars. As the scale of Blackwell products expands, gross margin is expected to slow to below 70%.
How does Wall Street look at this?
Wall Street institutions generally stated that nvidia will continue to maintain its leadership position in AI, as the industrialization of AI is still in its early stages. The company's latest AI chip, Blackwell, will continue to drive performance upward.
According to Citigroup's latest forecast data, by 2025, the capital expenditures related to datacenters of the four largest tech giants in the usa (amazon, google, microsoft, Meta) are expected to grow by at least 40% year-on-year. This massive capital expenditure is mainly linked to generative ai, indicating that the demand for computing power remains substantial.
Citigroup also pointed out that nvidia has the ability to capture up to 80%-90% of the market share in the datacenter AI server field in the future. The AI GPU, due to its hardware flexibility and broad adaptability to the iteration of large AI models, will become the preferred choice, with nvidia's ultra-powerful performance AI GPU + CUDA ecosystem creating an extremely wide moat.
After the performance announcement, many institutions raised their target prices for nvidia. Bernstein increased its target price from 155 dollars to 175 dollars; jpmorgan raised nvidia's target price from 155 dollars to 170 dollars; financial services company Stifel analyst Ruben Roy raised his target price for nvidia from 165 dollars to 180 dollars.
According to futubull data, the current average forecast for nvidia's target price on Wall Street is 169.78 dollars, indicating a 16% upside from the latest closing price.
Want to quickly find the key points of the performance? Open the search box >Performance Express, futubull helps you capture everything.
Editor/new
