Forget Nvidia. This Is the New King of AI

Hey everyone, welcome back to another video for today. It's a pretty good day if you're an AI investor. It's a very bad day if you're called Borito. Today we are going to talk about some AI names and some big news in the whole AI industry. Now, at the time of recording, of course, Core Reef and Nebus already reported their quarterly figures. Both of these names have had excellent quarters and more importantly, the comments, the forward guidance was very good. And that's why these names and all the other new clouds out there, but especially these names are up significantly today. Nebus is up 23% and Kore is up 17.7%. As always, Core gets a lot less love from the market as of right now. Let's see if that changes because I do think that this quarter was a very good quarter that showed that they are going into the right direction, especially especially when it comes to margins. As for Nebuse, again, we'll talk about that. If you missed the live stream, quick recap will happen in this video. Nebuse up 23% at $238 per share. Again, they are all doing extremely well. Now, a company that's of course central to all of these companies, especially a Nebus and a corewave, is of course Nvidia. And Nvidia year to date is now up. But for a long period of time, Nvidia was flat year to date. And that's that's quite strange because some people think, oh, Nvidia can be replaced. Nvidia's time near the sun is over. But I don't think I don't think that's the correct way to think about it. So I already apologize for the title. But when we look at Nvidia right at one point Nvidia was experiencing a draw down of 19% or so. Year to date now it is up 18.5%. But as you can see two weeks ago it was close to being flat year to date. And it's was close to being flat despite despite it trading at a trading PE of 34.3 times but a forward one of 22.4 times and even price to free cash flow is 22.7 times. I'll I'll remind you that Nvidia is a company that's valued at over $5 trillion. And a company that's worth more than $5 trillion is still expected to grow revenue by 82% this fiscal year, 42% in fiscal 28 and 22 close to 23% in fiscal 29. and look at the quarterly figures, it's it's even more astonishing because you would have assumed that the growth is going to come down. But no, this quarter they're going to report in 2 weeks or so. I think we're going to see a quarter where they grew revenue over 100% year-over-year, which means again an acceleration in growth. It's been a year of accelerating growth for Nvidia despite despite the revenue being 91 or expected to be over $91 billion afterwards. It's still 80% year-over-year growth 70 54 and 47. And I do also believe that these numbers are probably going to be beaten by Nvidia. If we look at their track record right from the beats consistently consistently beating analyst expectations analysts are expecting 91.8 billion wouldn't be surprising if we see 92 maybe even 93 billion. Last quarter 3.2% beat the quarter before that close to 3% then 3.7 1.5 1.9. So, we can assume that the beat is going to be closer, in my opinion, close to 3%. Now, that's not the only thing we need to talk about. We need to talk about this. This was a recent announcement. Nvidia partners with Apollo, Black Rockck, Blackstone, Brookfield, Goldman Sachs, and KKR to establish AI compute infrastructure financing platforms to mobilize over $500 billion of third-party capital. New financing platforms turn Nvidia compute and full stack AI infrastructure into an investable asset class for global capital broadening access to AI factories, enabling longduration usage link revenue while supporting Nvidia's ecosystem growth across hardware sales and software adoption. Of course, this is great for Nvidia because that means that well, Nvidia can continue and will continue to sell more and more GPUs to all of these companies out there. That also means you're locking customers in into their ecosystem into CUDA. Now moving a step back, AILABs, NeoClouds, well cost of capital now is expected to be lower, which means you can grow much faster. If you can grow much faster, you can have more compute come online. Although building data centers and all of these things, it doesn't matter how much money you have, things take time. But this will allow the whole ecosystem to just move much much faster. And well, yeah, again, the cost of capital, the cost of building all of it is in my opinion going to come down, which is of course great for the AI labs and Antropic and Open AI, SpaceX AI, but especially also the core, the Nebus, the Irons, and so on and so forth. Jensen said the following. Nvidia has reached an important milestone. We began by building chips. Today we are helping create a new asset class for productive investable infrastructure AI factories. In AI computes revenue, NVIDIA compute is uniquely suited for this role. It is broadly adopted, flexible across models and workloads, fungeible and transferable across customers and operators and continuously improved through CUDA software, extending its useful life and improving its economics over time. Of course, here is is slightly taking a little dig towards A6 because here they say look with us any company can use it, any model can use it. We're not building Nvidia GPUs specifically for specific workloads that we need. No, we're building it for broad adoption, right? Google TPUs are great, but they're mostly extremely good for Google's own use, which is is good enough. Same with Amazon. the usage there the business has been growing has been extremely successful but when you look at the adoption when you look at the ease of use it's quite clear that Nvidia here is still leading and I don't think that's going to change anytime soon so this was a very good announcement that we will see the impact of it I believe in 27 and onwards now looking towards 27 we might have another huge IPO coming up according to Wall Street Journal they say that entropic IPO could happen in October. The current run rate is $70 billion in ARR. They last raised at $965 billion. Probably going to aim for $1.5 trillion on IPO. Wouldn't be surprised if if it's that number or maybe they'll try to do a SpaceX number. Now, when this will happen, we might see two things. One, the same thing we've seen with SpaceX and space stocks, a huge runup in valuation until the IPO and then a crash. We've seen that before. Second of all, I do think that this is going to allow the market to get way more answers to look at an AI lab and see is this a profitable business? Is there a path to profitability? How fast are they growing? How much money are they burning? Hopefully, we do have or we do get a detailed report of each model. Which model is maybe more profitable than the other? Fable 5, Opus 4.8, sonet 5, etc., etc. So I am expecting a lot more volatility once an entropic and open AAI goes public. By the way, I said it a couple of weeks ago. I think if they were public in July, these names would have been cut by 50% easy. But if they go public later this year and the market comes down because we have way more information and the information is quite good. I think it's going to be very very healthy for this whole ecosystem. Now moving on to of course my two favorites here, Nebus and Core Reef. Nebus is my favorite. These earnings reports were really, really good. So, let's dive into them. If you enjoy this type of videos, hit all the buttons. We really appreciate that. Want to support me even further, do check out the link down in the description and in the pin comment to the top 10 best stocks to buy now or go to full.com/cashinves investor. Thank you very much. So, on the left here, we have Nebus revenue there grew 454% year-over-year. Of course, way smaller base than a core, but it's growing fast. It's at $582 million. ARR is up to $3 billion. That's up close to 600% year-over-year. Cloud profitability 50% AI cloud adjusted Ebida margin $236 million for the group total. They of course announce an asset glide capacity model and short-term premium tiers that they see 40 to $50 million per megawatt which is insane. We'll look at what that might mean for 27 and onwards. And by the way, the short-term pricing and contracts, we have also seen that with a core reef. Why did they do this? It's purely because they are getting a huge premium. As for the financial payback, this is also huge for Nebus. They compressed deal payback to 1 year and 10 months, which previously was I think two to three years or so. And I do believe that number could come down drastically. As for Coreweave, revenue is up 112% year-over-year to $2.57 billion, up 24% quarter over quarter. Revenue backlog 104.2 billion. But they did already tell us that they signed $25 billion early in Q3. As for adjusted EIDA, that sits at 1.51 billion, 59% margin versus negative $626 million of a gap net loss. Nebus told us that they closed major deals with reflection coher and an unnamed US NEOLAB and a top US quant trading firm. We don't know who that is. The average TCV exceeds a billion for these deals shifting ACV 2 megawatt from $12 million base to 20 to $25 million. So that's already a big jump. And then they're already seeing another big jump to 40 to50 million. Moving on, they had their first capacity auction. What did they see? only positive things. They priced 15% above any previous Blackwell high ever cleared in the market. They introduced an explicit live price discovery engine to capture acute market uh demand. And even the customer that paid that price was extremely happy with it and wants to of course get more. As for the short-term premium tier, they signed the first three to six month premium capacity contract in Q3 that's going live in Q4, unlocking an unprecedented 40 to $50 million per megawatt pricing for immediate need workloads. Now, the asset light business or partnership model was already announced I believe a month or so ago. So, the partner provides land and facility, power infrastructure, hardware, capital, and facility ops. Nebus will provide the full stack software system architecture, global grow to market and demand and of course this converts the partner capacity into high margin ARR with minimal Nebus capex which means that this will just accelerate Nebus business going forward. They now expect to end this year with 5 gawatt of contracted power. Previously that was over 4 gawatt. With regards to the violent situation, they told us the following. By the way, violent capacity is a 2026, not a 2027. They did say the following thing. They switched this site layout to Bloom Energy fuel cells providing quiet, reliable, ultra- low emission power directly on site. Trenches delivered on schedule, building finished earlier this summer. Engineering fit out progressing well. And as of right now, they still believe everything's going to end on time. And here's the thing. They told us that connected power for this year or this year end is still expected to be between 800 megawatt to 1 gaw. And then from my understanding they said during the call that this could then translate into active power mid 2027. And so we could we could potentially get an ARR number that sits close to 20 to 25 billion for 27. Of course, mid27 is already 6 months into the year. But it wouldn't be surprising to me if we are going to have a 20 to25 billion ARR number somewhere in 2027, which is way higher than what the market is thinking about right now, right? Cuz we went from the base pricing of $12 million per megawatt. Then we're at 20 to$25 million per megawatt. And then we got the premium short-term contracts where they're already getting $40 to $50 million per megawatt. And so yeah, I do think this huge jump is not priced in as of right now. Moving on to core wave, we talk here about uh the debt and of course the interest drag. Now they did raise $18 billion in Q2. There's over $32 billion of total capital. Now the weighted average debt cost was reduced by around $300 basis points year-over-year, unlocking around $1.1 billion in annualized interest savings. It is a pretty big deal because as you know in Q2 interest payment was interest expense was $640 million more than doubled year-over-year and the guide for Q3 is expected to be between $860 to $940 million. Of course, there is a reason for all of it. They are growing like crazy. They have now over 1.5 gawatt of active power. They've added around 500 megawatts of active power in Q2 alone. the largest quarterly expansion in company's history. They have now 51 data centers and they have 1 gawatt contracted outside the United States including a 360 megawatt entry in APAC in Indonesia expected to be online in around 18 months. As for the added services, in this case the manage inference business, that grew from just a million dollars to over hund00 million in ARR in a single quarter, and they're guiding for that number to reach over $250 million in ARR by the end of the year. NonGPU software storage services also exceeded $400 million in ARR. They also implemented a 25% list price increase across SKUs in July. Q2 contracts carry around 5 to 10 points higher contribution margins especially Vera Rubin clusters and then lastly they signed a 2020 A100 GPU contract extending out to 2029 proving that long-term value retention and monetization across legacy GPU fleets. And so all in all when you compare both of these quarters and the guidance it's it was again a very very good quarter. The momentum is definitely there with these players and you need to understand that this is a market and industry where yes there is a lot of money to be made but they are building huge huge buildings that is very very complex the GPUs the cooling all of it the networking all of it will take time right you cannot go unless you're Elon and SpaceX AI who thinks they can go to 10 GW in a year or so close to it I'd love to see that that would be insane and if they can do it. Maybe others can do it as well. But what you're seeing here today, the Nebus revenue, the core reef revenue, it's going to be looking extremely extremely small a year, two years from now. As for the fullear guidance, Nebus basically said that everything stays the same. They reaffirmed that. So capital expenditures capex here is still 20 to$25 billion. I wouldn't be surprised if that number doubles in 2027. Cororweift did already raise it this year, this fiscal year to 35 to 39 billion. But then again, uh the active power number also increased. So that makes a total sense. Full year revenue is expected to come between 12.4 to 13.2 billion with an 18.5 to 19.5 billion exit run rate. And so when we listen to all the naysayers, when we listen to the shores, when you listen to the bears, to the people that don't really understand what's going on here, or maybe to the skeptics, right now the earning season is almost over. We've had all of the big tech companies, the hyperscalers report, give us guidance. The momentum is there. We now are having the Neocloud's players say, "Look, the momentum is there. The results are here. We are all expected to grow much more in the future." By the way, Nebus is expected to add a gigawatt every single year. starting in 2027. And so, yes, we are waiting for probably Oracle and an Nvidia. Those will happen later this month. But for now, there's nothing that tells me that one, there is a bubble, and two, that this possible bubble is going to pop anytime soon because the growth is there for some of these players. The profitability is there already right now. And so, yeah, this is just amazing, amazing to see. Congrats to all the shareholders. And in my opinion, Nvidia can continue to crush it. No question about that. So, all in all, that's about it for me in today's video. If you enjoy this type of videos, hit all the buttons. We shall see each other in the next one. Bye-bye.