NVDA收购Hugging Face旨在部署多余的租赁计算能力并强化CUDA护城河；然而，从数据中心（营业利润率71%）向企业/图形（营业利润率49.2%）的混合转变可能威胁利润率。

Uh look at this though. AMD now trying to p sorry Nvidia trying to push into uh the 220s. Really I think Dell is this signal that hey the hardware rally it ain't over yet baby. It ain't over until it's over. My hope uh is that we get our hardware 2.0 0 rally once we're done with a whole set of earnings and you can get Nvidia actually finally sustain some new highs, you get new highs across hardware.

There's Nvidia's still pushing though. Those Dell numbers, man. Look at that. Nvidia is having the opposite impact.

Nvidia is holding on to two and a half as well. Micron, Nvidia, Broadcom.

Um now post Nvidia's earnings where Nvidia shocked everyone with that outlook for fiscal 28 on growth >> the 70%

And it was a very interesting moment because there are all of these deals that have been announced. The TPU with Google is probably Broadcom's most best known project, but there are others out there. And so scaling those, putting them out into the real world, people are looking for evidence, not only that that's happening, but that the chips actually are good, they work. And you guys will remember back to Monday when we had Jensen Wong on the show. In open interest, he could not have been more clear. Like they acknowledge that custom silicon is a factor in this market. Um, how good it is relative to Nvidia is important, but also how widely deployed they are at scale is a bigger consideration to Jensen's mind as well.

>> And you know, it's been interesting because you pointed out Nvidia really just injecting fuel into the optimism of AI and the uptake of it with that outlook.

Oh, yeah. Here we go. Oh, and there's Jensen with Lutnik as well. That could be interesting. >> layers. um you know they uh Dar Koshi the CEO talked about reducing micro teams you know teams that have one to two reports >> yeah we talked about that already and covered that okay let's go see if there's anything juicy in this

about how important the industrial new industrial revolution is nothing exciting >> around these AIs the environment around these AIs so that they could do productive work all of that harnessing all of that environment creation is what companies do is what leaders do as it turns out our jobs won't change our jobs is defined by purpose context meaning that's what a job is in that job there's a lot of typing and talking and you know things like that those skills are going to be automated by AI the tasks are going to be automated but our our jobs purpose remain and so I think >> purpose remains good thanks Jensen

>> purpose remains good thanks Jensen >> ambition of my colleagues my own ambition has really been supercharged and so >> we've heard a lot of these before I'm going to do like one more clip here and then I'm going to go the Sam Alman one which should be pretty interesting >> our countries and benefit society and I think what you've done uh Jensen which we really appreciate I mean you are an American treasure >> right your company uh is leading us in a path that none of us when we were young ever dreamed possible maybe we dreamed it a little bit that's what you know sort of sci-fi was about but now we're there I think what you've laid out for us is a vision vision of that harnessing AI to using it both from uh surrounding it, contextualizing it and directing it is how we're going to succeed that each of the countries here needs to invest in its infrastructure in order >> exactly. Okay, got it. So, the Jensen piece me not that much uh entertaining color.

Ian why would Nvidia want to buy hugging face as Rachel's just described it >> what she said was key there which is an open models, right? At the moment, the industry is very concentrated on the hyperscalers and what they produce, on what Open AI does. >> And what Nvidia really wants is a broad adoption of AI across the economy. It basically wants more customers.

I think it sounds a little bit surprising on on its face, but then if you think about it, it actually could make a lot of sense. Uh the company has it's been around for about 10 years. They have a lot of investors including Nvidia. Um my understanding from people that I've spoken to um that are close to the company say that Nvidia has been long been a dedicated investor of the company and as Ian mentioned, they're aligned um especially on the open model.

Bit of M&A history for Nvidia, Ian. You know, we say $14 billion would be like their biggest acquisition in AI. Where has Nvidia done actual acquisitions as opposed to like these investments which it makes many of? >> Yeah. I mean it doesn't have a huge record of acquiring it. It tends to like to do things in house. You could probably the most recent one of note was what it did in terms of getting hold of the assets of Grock the >> Grock with a Q. >> That's right. Grock with a Q, the AI chip investor.

the story on the technology side has been that Nvidia dominates the content ownership of the server. You know, Dell doesn't really contribute that much to it.

And and Nvidia's trading on about 10 times PE if you go out a couple of years. And so, you know, what did Tim Cook do as soon as he kind of got the reigns? He he put in a massive shareholder program to say, you know, we're not just selling a a phone. We're not the next Nokia Motorola. This is sustainable and we're going to prove it by giving you, you know, one of the the greatest shareholder return programs in the history of US corporate America. And it's the same with Nvidia. They're embarking on that journey as well. So, you know, that's going to be the only way that there's only limits to what they can do in terms of convincing people this is durable and this is going to last beyond next year, which is currently what the valuation is telling you the market doesn't believe.

because we're seeing you know big buybacks from from Nvidia because the free cash flow has dried up for many of those companies those buyback programs have gone away

You know to me that would sound like uh AMD, Nvidia, right?

Oh, wow. Look at Nvidia, dude. Nvidia just rocketed. I think this is on the back of Dell, honestly, just the continued hardware spend. But wow, look at that. Straight to 227. That's that's good. You know, take take hardware with it.

Nvidia Hugging Face is the merger and acquisition to watch. There's really no news other than Nvidia said to near 14 billion deal on Hugging Face and that was from this morning.

Okay, so there's some there's some talk that maybe Nvidia can use the acquisition to basically tilt distribution towards CUDA programming and they might be able to steer compute to the more Nvidia friendly ones. That's pretty jaded, but that's fine.

My point of view is that they're acquiring hugging face because they keep, you know, signing lease contracts and they need to put a purpose to that compute. So, both of them are kind of jaded outlooks.

We signed data center lease agreements with with terms of approximately 15 years. I think that's why they need hugging face because, you know, they can uh they could get to work on deploying the very own compute that they're buying.

and on the back of Dell, I think some people see Nvidia as this, you know, if we look at Nvidia as, you know, it's it's both the inference and the uh training chip winner. Uh, you know, that's at least the thesis. Although, you know, compute uh has lower margins. We know that. We should be able to actually see that uh that's probably going to be their actual earnings though.

Farah Rubin now in full production margins. Margins margins are going to be down here. Nvidia investor relations. Okay, here we go. Financials SEC filings 10 Q's out. Let's go.

Okay, here we might see some more of that leasing commentary lease. Oh man, 114 times goddess center leases not yet com commenced. What is this? This is primarily compute chips network expected to begin between the third. So it's actually pretty nominal like compare. I mean but these are in billions, right?

Yeah. Yeah. Yeah. Yeah. See, they're purposefully showing this in billions to make it seem like it's nothing, but it's still a commit. Wow, look at those commitments. Oh. Oh, but you're adding together like potentially 15 years worth. So, you know, that could be $1.4 billion, $1.4 billion, and then it rounds up. So, that's interesting. That's the leasing side.

Uh, leases. Okay, so we got the leases. Now what we need is there a little bit more color but these are and these include offices and data centers blah blah blah blah blah. What about segments? Segments compute and networking graphics.

No, I want to see like direct customers. Where do they put the 5090s like the the gaming segment? I think they've consolidated that now. There the graphics segments includes GPUs for workstations. See right here works enterprise workstation graphics. Okay, so this is where you know workstations equal graphics.

And now we can go to segment operating income. So now we're going to go to graphics and see what the margins look like. Graphics. Here we go. Operating income. Huh. Oh, perfect. So that's the operating income. Here's the revenue. So graphic revenue graphic rev and then here we have operating income graphic income.

So graphic income 3899 divided by 7922 49%. That's weak. 49 compared to their other operating income. Operating income is it though? Because maybe I'm thinking let me well I guess we'll find out but I'm pretty sure that's weaker than than compute and networking because that's where the data centers are. So here we have 62696 divided by 88299 equals Oh yeah, dude. 71% going to operating income. That's a huge difference.

Op uh operating income from data centers and this is operating income from workstation segment bundled together with you know some of the other crap but that's worth paying attention to. Okay. Okay. Uh, look at that, man. 227 hanging on.

You know what's his name? Um, George Gam had a piece on uh comparing Nvidia to Cisco. Uh, and you know, there are some comparisons that can be made, but it's still so early.

We'll come back to Nvidia. There's a lot to catch up on and there's even, you know, cyber cap tomorrow, some talk about the data center spend. The Sam Altman commentary that was really interesting. So, we got a little bit of catching up to do.

And then we've got the Nvidia risks to talk about.

Okay, here goes Nvidia losing that line. Yeah, you really struggle with that line.

Jensen Huang, I I caught up with him from Nvidia on the sidelines of the summit this morning and he talked about regulating actual risk but not hypothetical risk and so making sure that they're allowed to keep going before there are more rules in place.

And where's Nvidia? Yeah, Nvidia's right here. Ah, I lost that 227 again. Come on, man. Dell is supposed to push you to the new levels. It's all right.

this hardware issue with uh the CPOS's, the Nvidia, the Dell, the Sam Alman, the Core Weave. All of this kind of loops together.

because it helps us explain potentially why we are seeing some shifts in stocks including Nvidia up 3.7% today.

Nvidia is probably popping today on this reportedly nearly complete $14 billion acquisition of Hugging Face. It's about a 12.9 billion acquisition plus about a billion dollars in employee retention. You know, got to give got to give the employees a billion dollars just to stick around.

What Sam just said, Nvidia's acquisition of Hugging Face, and the Neocloud issue. So, how do we pull this together? Well, first we have to remember that Nvidia has got 15-year lease commitments for some uh data centers uh or or for some compute that they're trying to actively pawn off on third parties. They say this here in the CFL letter, "We signed data center lease agreements with terms of approximately 15 years that are expected to commence between 2028 and 2029 uh fiscal year. Uh and we expect to reassign those data center leases to third parties. So given that we're in fiscal 27 right now, this basically means these leases are starting next year and the year after that. They're trying to pawn them off on third parties, you know, maybe private credit, but they're signing these contracts here saying, "Hey, we'll lease a bunch of compute."

A jaded view of the hugging face acquisition by Nvidia is that they have all this compute that they've just secured and that maybe there is a little too much compute being built. And so Nvidia is coming in swooping some of this surplus compute. One of the companies they bought surplus compute from or promised to buy surplus compute from was Coreweave and they said, "Hey, let us know if you ever need us to buy your surplus compute." And then they did. Coreweave said, "Yeah, we've got surplus compute. We can't get rid of Nvidia. You said you would guarantee it. You would buy it. Here you go. Buy it." So now Nvidia is like, "Great. We just wanted to sell the chips. Now we have compute that we don't really need." What if we bought a business that needs compute? Huh? Welcome to Hugging Face.

Some jaded views are that Nvidia oversimplifying here but that Nvidia can now utilize this power of controlling hugging face to guide model adoption towards CUDA or Invidia compute and potentially enable or support that very compute with the data center leases they have now secured. So, in other words, you have this complete circle. Uh, and maybe it's a jaded point of view, but maybe that's exactly why Nvidia is going after hugging face.

Well, now Nvidia is like, "Great. We were not expecting to have to take that compute so early. So, we need to put this to work. So, what can we do?" Well, let's buy something that needs compute or could utilize compute. Why don't we buy a software company like hugging face and we can put that compute to use. Now all of a sudden you can kind of compute complete the cycle where the additional jade comes in that says if you could then guide models to be a little bit more CUDA friendly which is that uh you know programming layer on the uh graphic cards that Nvidia uses massive moat uh in the artificial intelligence space then you could potentially drive more compute demand to the neoclouds and kind of complete that circle anytime they have excess capacity Nvidia will find a use for it.

It's obviously weird that they talk about trying to offload this risk to third parties and it suggests that you don't really want this exposure. And I don't think it's a surprise that in their 10Q filing, they kind of purposefully downplay these leases. See, when you look at this chart, they have this chart written in here as they purposefully put data center uh supply and capacity. In other words, like, hey, uh, this is what we're promising to still supply. These are our supply commitments. Increasing our commitments for critical components for data centers. We're saying this is how much we're still going to spend on core components, high bandwidth, memory or whatever going into these chips. Then they purposefully try to bury this right here. Data center leases not yet commenced. These are those leases that they're committing to over a 15-year period. And because they can write this in billions, because they're comparing to their commitments, it doesn't actually look like that big of a deal. But starting next year, they literally have a one over $1 billion. We don't know because there's rounding because again, they're showing this in billions. They have over a billion dollars in data center leases expenses coming next year. That's a lot. So, they need a purpose for that. Again, this is the only chart they really show in billions uh that that where the numbers get so tight and small and it's fine. I get it. They've got a lot of big revenues. It makes sense. But a lot of other charts like we go over here and we want to look at, you know, gains that they have. Oh, what a surprise. These are being shown in millions. So, why are we trying to minimize those leases and make them hide? You know, we go to the cash flow statement. Oh, wow. It's shown in millions, but those leases are shown in billions because it makes the numbers look a little smaller.

because we have to somehow consider those data centers might need Nvidia to buy back some of the data centers.

Well, then we don't need to pay for the GB300, the GB200 or H100s or other Blackwell chips. We don't need these. We don't need the Vera Rubin for that enterprise inference.

But that's also where Nvidia rejecting 227 again today is interesting because a lot of people say, "Okay, so Nvidia wins either way. Nvidia wins with Frontier Compute. Nvidia wins with uh Enterprise Compute, does it?" Well, that's also where things are a little blurry.

See, if you actually look and break down uh the graphics income uh which is exactly where Nvidia says that the graphics segment includes GPUs for gaming and enterprise workstations. So the Blackwell 6000, the 5090, whatever those are included in graphics. Graphics revenue is broken down right here compared to data center revenues, comput and networking right here. We can compare that to their operating income for those segments and we can come up with margins for those segments. We can see that data centers for Nvidia are pushing a 71% operating income margin, which obviously explains why Nvidia really wants to sell more data center chips, which explains why they might want to buy HuggingFace to kind of keep the NeoCloud supported and keep those leases operating because they want to keep selling data center chips because they've got a 71% operating margin on that. Really, really good margin. the operating margin from the graphics side, so gaming and workstation AI and more enterprise AI only 49.2%. Still really good, but a fraction of the 71%. I mean, in fairness, nearly 50% operating income is still very good. It's just a whole lot lower than 71%. And when Nvidia forecasts even like a 1% miss on gross margins, people freak out and the stock ends up falling. So the margins matter a lot to Nvidia.

And so if you shift to enterprise compute, you're probably going to see more margin weakness at NVIDIA in the future. If you shift away from the hyperscalers and into enterprise compute, it's not margin accreative to Nvidia. It hurts margin for Nvidia.