In essentially the most anticipated quarter of this reporting season, Nvidia far exceeded high expectations for revenue and profit. Even higher was a giant revenue forecast and a broader vision from CEO Jensen Huang that bolstered the notion that corporations and countries are working with the AI chip powerhouse to remodel $1 trillion of traditional data centers to accelerated computing. According to data provider LSEG, formerly often called Refinitiv, revenue in the primary quarter of fiscal 2025 rose 262% 12 months over 12 months to $26.04 billion, well above analysts’ forecasts of $24.65 billion. Dollar. The company had previously forecast revenue of $24 billion, plus minus 2% – in order that was an enormous increase. Adjusted earnings per share rose 461% to $6.12, beating the LSEG-compiled consensus estimate of $5.59. Adjusted gross margin of 78.9% also beat Wall Street’s estimate of 77.2%, in keeping with market data platform FactSet. The company had forecast gross margins of 77%, give or take 50 basis points. In addition to the strong results, Nvidia announced a 1-for-10 stock split. Although stock splits don’t technically create value, they typically have a positive impact on the stock. The company said the split was intended to “make stock ownership more accessible to employees and investors.” We commend Nvidia for this motion and can proceed to pressure other corporations to do the identical. Nvidia last split its shares in a 1:4 ratio in July 2021. In after-hours trading, the rapid rise in Nvidia shares was not a giant surprise. Nvidia Why We Own It: Nvidia’s powerful graphics processing units (GPUs) are the principal driver of the AI revolution, powering the accelerated data centers being built rapidly world wide. But that is greater than only a hardware story. With its Nvidia AI Enterprise service, Nvidia is within the strategy of constructing a potentially huge software business. Competitors: Advanced Micro Devices and Intel Last purchase: August 31, 2022 Launch: March 2019 Conclusion Which air hole? At the beginning of the quarter, it seemed like the one thing holding Nvidia back was a product transition-related slowdown, as customers delayed their orders of the H100 and H200 GPUs (graphics processing units) in anticipation of the superior Blackwell chip platform. As you possibly can see from Nvidia’s big beat and upside guide, this has been removed from the case and demand is anticipated to proceed to outstrip supply for quite a while. If this narrative repeats itself, here’s what it’s best to remember next time in order that these concerns don’t dissuade you from a powerful long-term thesis: Jensen explained on the earnings call that customers are still at such an early stage of their build-out imagine they should proceed buying chips to compete in the present technology arms race. And technology leadership is every little thing. “There’s going to be a whole bunch of new chips coming their way and they’re just going to have to keep building and, if you will, increase performance through averaging. So that’s the smartest thing you can do,” the CEO said. In general, we didn’t hear anything on Wednesday evening that may change our long-term view that Nvidia is the driving force behind the present AI industrial revolution. This is how Jensen explained the change happening: “In the long term, we are completely redesigning the way computers work. And this is a platform change. Of course, it has been compared to other platform changes in the past, but time will clearly show that it is much, much deeper than previous platform changes. And the reason for this is that the computer is no longer just a command-driven computer. It is a computer that understands intentions.” Jensen further mentioned that computers not only interact with us, “but also understand our meaning, what we ask of them, and that they have the ability to reason, to reason iteratively to process and plan and come back with a solution.” The billions and billions of dollars spent on accelerated computing are why we own Nvidia for the long term and don’t try to trade the company back and forth with every headline . By the way, another pessimistic story we often hear is that the custom chips that all the major cloud companies make pose a threat to Nvidia’s leadership position. Jensen doesn’t see it that way, as his platform system offers the highest performance with the lowest total cost of ownership. That’s an unbeatable value proposition. NVDA YTD Mountain Nvidia YTD The strong results and outlook, upbeat commentary and stock split pushed Nvidia shares up about 6% to over $1,000 per share for the first time. However, we don’t think the profit margins end there. We raise our price target from $1,050 to $1,200 and maintain our rating of 2, meaning we view the company as a Buy on any dips. Quarterly Results Growth was driven by all customer types, but enterprise and consumer internet companies led the way. Large cloud companies accounted for about 40% of data center revenue in the quarter. So when you see companies like Oracle and Amazon and Microsoft and Alphabet raising their capital spending forecasts, expect a lot of those dollars to flow to Nvidia. And there is a good reason for that. In the call, Nvidia CFO Colette Kress estimated that for every dollar spent on Nvidia’s AI infrastructure, a cloud provider has the opportunity to earn $5 in revenue from instant GPU hosting over four years . One of the customers that called in the quarter was Tesla, which expanded its AI training cluster to 35,000 H100 GPUs (graphics processing units). Nvidia said Tesla’s use of Nvidia’s AI infrastructure “paved the way” for the “game-changing performance” of fully autonomous version 12. (Fully autonomous driving, or FSD, is the way Tesla markets its high-end driver assistance software.) Interestingly Nvidia sees the automotive industry as a huge vertical market this year, a multi-billion dollar revenue opportunity across on-premise and cloud usage. Another highlight was Meta’s announcement of Llama 3, its large language model. It was trained on a cluster of 24,000 H100 GPUs. Kress believes Nvidia will see more growth opportunities as more Internet customers adopt generative AI applications. The Tesla and Meta clusters are examples of what Nvidia calls “AI factories.” The company believes that “these next-generation data centers house advanced full-stack accelerated computing platforms where data comes in and intelligence comes out.” Nvidia also identified that sovereign AI is a serious source of growth. The company defines sovereign AI as “a nation’s ability to produce artificial intelligence using its own infrastructure, data, workforce and business networks.” Kress expects revenue from sovereign AI to succeed in high single-digit billions this 12 months, after failing to succeed in it last 12 months. Looking forward, Nvidia expects supply to enhance with the H100, but continues to be limited with the H200. Even with the transition to Blackwell, Nvidia expects demand for Hopper to proceed for quite a while. “Everyone is eager to get their infrastructure online and the reason for that is because they save money and make money, and they want to do it as quickly as possible,” the corporate said. In other words: customers take what they will get. But expect revenue from Blackwell later this 12 months, possibly in a really significant amount. The company said Blackwell production has already began and deliveries are expected to start within the second quarter of fiscal 2025 and ramp up within the third quarter, with customers having full data centers within the fourth quarter. As for China, the corporate said it has begun ramping up recent products made specifically for the region that don’t require an export control license. The US government has imposed restrictions on the sale of the fastest chips, fearing they could possibly be utilized by the Chinese military. However, China just isn’t expected to be a revenue driver because it has been up to now, as limitations on Nvidia’s technology have made the environment more competitive. Forecast The company’s fiscal second-quarter guidance should allay market concerns that some form of “air hole” may be forming for AI spending. For the current second quarter, Nvidia forecast revenue of $28 billion, plus or minus 2%, above consensus estimates of $26.6 billion. Adjusted gross margins are expected to be 75.5%, plus or minus 50 basis points, above estimates of 75.2%. Return on Capital Nvidia increased its quarterly dividend by 150%, which is nice, but the annual yield is insignificant to the investment. More impactful is the $7.7 billion in shares the company repurchased in the fiscal first quarter. (Jim Cramer’s Charitable Trust is long NVDA. See a full list of stocks here.) As a subscriber to CNBC Investing Club with Jim Cramer, you will receive a trade alert before Jim makes a trade. After sending a trade alert, Jim waits 45 minutes before buying or selling a stock from his charitable foundation’s portfolio. If Jim talked about a stock on TV, he waits 72 hours after the trade alert is issued before executing the trade. THE INVESTING CLUB INFORMATION ABOVE IS SUBJECT TO OUR TERMS AND CONDITIONS AND PRIVACY POLICY AND OUR DISCLAIMER. NO FIDUCIARY OBLIGATION OR OBLIGATION SHALL EXIST OR CREATE BY RECEIVING ANY INFORMATION RELATED TO THE INVESTING CLUB. NO PARTICULAR RESULT OR WIN IS GUARANTEED.
Jensen Huang, co-founder and CEO of Nvidia Corp., during the Nvidia GPU Technology Conference (GTC) in San Jose, California, USA, on Tuesday, March 19, 2024.
David Paul Morris | Bloomberg |
In essentially the most anticipated quarter of this earnings season, Nvidia far exceeded high revenue and profit expectations. Even higher was a giant revenue forecast and a broader vision from CEO Jensen Huang that bolstered the notion that corporations and countries are working with the AI chip giant to remodel $1 trillion of traditional data centers to accelerated computing.
