1. NVIDIA Corporation (NASDAQ:NVDA)
Number of Hedge Fund Holders In Q4 2024: 223
NVIDIA Corporation (NASDAQ:NVDA) is Wall Street’s favorite AI stock, and it continues to reflect investor sentiment about the technology. While its shares had remained mostly flat during the second half of 2024, things took a turn for the worse in 2025 when NVIDIA Corporation (NASDAQ:NVDA)’s shares dipped by 17% in January during the DeepSeek selloff. The shares have been down by another 10.8% since its latest earnings, which is the back of fresh worries about additional sanctions on the firm’s GPU sales to China. Cramer’s latest remarks for NVIDIA Corporation (NASDAQ:NVDA) were made after the earnings:
“[On Goldman saying NVDA’s earnings were good enough] Yeah, I mean Ben Reitzes, it was, we were going back and forth this morning, and you can say, look, it was a quarter, frankly, GPU business and the GPUs number, were beaten. I think that one of the difficulties of understanding, it was the whole China stuff. . . .what matters is that, the Chinese initiative is child’s play. DeepSeek can’t do it. I think everyone has to start thinking, what Blackwell does, is it reasoning? So if you go on to ChatGPT right now and you say, is Walgreens going to get a bid from Rob Sweeney at Sycamore. It would say, Sycamore is a company that does retail, Rob Sweeney used to work at Goldman Sachs. Uh, Walgreens is run by Tim Wentworth. And you’ll be like, well that’s . . .worthless. Well, what he’s talking about, it’s going to be reasoning, it’s going to say that’s a possibility, it might happen. The balance sheet is not that good.”
“I’m just saying, okay that, it’s maybe, John did a great job. It’s really important to try to figure out what happened with the narrative. When I speak with the company, the narrative was, frankly they were shocked that China, the China DeepSeek would have played any role given the fact that all the hyperscalers have placed all these orders, they all knew about DeepSeek. They’re much more into another world, post-training world, they’re literally in a world where, they want to have robot. Robot comes into a warehouse, and the robot looks at the warehouse and says you know what I gotta put this on a truck. I have to go around this corner, I have to lift it up. I gotta be very careful, I don’t want to damage the box. Whereas right now the robot says tell me what to do. The next robots says, okay, here’s how I’m going to do it. And that’s where we are. And that takes it to a whole different level of compute. And it’s, Jensen keeps using the term compute, meaning that look we don’t just, remember they’re both accelerated computing and generative AI. The compute is so fast that they can actually do things. Right now, Jensen’s famously saying it’s so slow. I mean to us it’s fast. But it’s a parlor game to him. And this is, something that’s industrial and will replace all of every bit of computing we have right now.”
“[On whether the margin guidance is good enough] I think it’ll be even better than that. You know if you go back, one of the ways to look at it, this is one of the ways that Jensen told me. If you go back to last year, at this time, this week, and you looked at what the stock traded at on forward earnings, it traded nineteen times earnings. The stock always looks expensive and then when we get there the stock turns out to be very cheap. So I could argue if this stock, if you give it a, you can put a higher multiple on, you can get to one fifty, one sixty.”
“[Citi’s price target being 23x CY26 earnings] Okay, that’s a very wise person. That’s exactly what I’m saying. If you look at it last year you would realize holy cow this stock turned out to be very cheap. Now this is where the hiccup. This is not, you know, this is not smooth thing. The number of factories that are involved, the number of companies involved, the number of touches. This is thing is monumental and it’s obviously very big. And Jensen’s still able to pull it off. I regard it as a great quarter. I thought it was a great quarter.”
“I just think you have to keep coming back to the software in the stack. And you have to come back to the accelerated computing. No one’s even near, as close, 25 times, they think that, DeepSeek is 125th of the speed and power of Blackwell. And Blackwell by the way, they have the next iteration coming later this year. And Rubin next year. And this is going to be a much smoother ramp. Because it’s the chassis is the same. So anybody who thinks that, because I asked them that, guys, after what happened, isn’t it, just could be a disaster right? And he said no, no. We have to get the chassis right.”
“Well the hiccup is they moved after one particular chassis . . .to another, and it was very, very hard. . . .But everything else is going to be based on that same chassis so you’re not going to have the same technical difficulties. Look, if you think about the fact that the, they’re talking about a lot of industrial uses. They’re talking about omniverse, to, omniverse is difficult, you have, you build a small computed factory. You make all the changes in the factory, and then you build a new factory. That is one use that they’re really highlighting. They’re highlighting healthcare uses. What’s lacking is they did not talk about use cases. That’s for GTC. That’s their, the Woodstock of AI. So otherwise you’re kind of left there thinking, what’s going to happen?”
“. . .They kept talking about speed, speed, speed. No one’s listening. We’re all listening to whether China’s going to get the orders, and China’s going to beat us. And what’s the President going to do. But the issue here is if we can get speed, then we can do something like a baby’s heart. I’m just giving you a use case that is so insane! But you have to think, we can all just say it’s not a Microsoft.”
“[On how the human brain cannot process an exponential rate of change] And that is exactly where Jensen is that we’re not. He can think of these things. I know that it sounds ridiculous. . . look I don’t want to say that we’re pedestrian thinkers. I will say that I am not able to comprehend things that Jensen does. . .You need so much compute to be able to do a heart!”
“If you aren’t a big thinker, I’m a pedestrian thinker, I can tell you whether NVIDIA’s going to go to one thirty-five versus one thirty-four. I cannot tell you what I would do if I had compute power. Like if you go to Jensen’s office he’s got that picture of a, of a rocket going basically going how fast it has to go to hit Mars. And he’s trying to figure out at what point it has to go to one mile and then reverse. And I said what are you doing with that? And he goes, I’m just calculating how quick it has to go.””
NVDA is a stock Jim Cramer recently discussed. While we acknowledge the potential of NVDA as an investment, our conviction lies in the belief that some AI stocks hold greater promise for delivering higher returns and doing so within a shorter time frame. If you are looking for an AI stock that is more promising than NVDA but that trades at less than 5 times its earnings, check out our report about the cheapest AI stock.
READ NEXT: 20 Best AI Stocks To Buy Now and Complete List of 59 AI Companies Under $2 Billion in Market Cap
Disclosure. None. Insider Monkey focuses on uncovering the best investment ideas of hedge funds and investors. Please subscribe to our daily free newsletter to get the latest investment ideas from hedge funds’ investor letters by entering your email address below.