Projects
![]()
No projects found
Investors
New
DAO Treasuries
DATs
Lend / Borrow APRs
Chains
Categories
Spaces
NVIDIA $NVDA LIVE EARNINGS | STOCK MARKET TALK
Recorded: Aug. 27, 2025 Duration: 3:15:18Space Recording
Full Transcription
Thank you. what is up what is up i was excited nvidia live earnings day so i decided to start the space a
couple seconds early i'm gonna get the title changed all that good stuff but i appreciate
all of you for joining we're gonna have a fantastic conversation nvidia earnings coming uparnings coming up 4.20 p.m. Eastern,
followed by the Earnings Call at 5 p.m. Eastern.
Jensen Wong also does a lot of interviews as well.
So expect a lot of Mr. Wong over the next three to four hours.
But let's get this title changed.
NVIDIA Live Earnings.
Stock Market Talk after.
All right. Title has been changed let me also go in and get my account in here how is everybody doing though i would love to hear what you guys are looking forward to on
these nvidia earnings well we'll talk plenty about it we already have 24 of you guys hanging
out within here throwing that in the comment section,
what you guys are hoping for on these NVIDIA earnings,
watching, don't want to hear.
How do you expect they're going to go?
I just want to hear your thoughts on the NVIDIA earnings.
Cool.
My account is coming up.
I'm looking at the market in my portfolio.
Apple is green, although the rest of my portfolio is red.
BMNR, we'll give it a nice shout-out nice and early.
I've been buying a little bit.
Who knows?
Clearly, I'm wrong today.
It's down 8% at $45.92.
BMNR is moving lower today.
Apple, though, is green.
NVIDIA, the big name on the day.
Earlier, we were coming in.
NVIDIA was starting the day up around,
in pre-market, it was up like a percent.
Not really got that far,
but we are hanging out around even on Nvidia 182.
That gives it a market cap of $4.44 trillion.
The implied move means we're about to have
like a $250 billion move on Nvidia
is what the market is sort of expecting in.
We will see on that.
But looking at the indexes, SPY is up 0.2%.
Where are Qs?
Qs are up.
Well, VOO is up 0.3% and QQQ is up 0.2%.
So the indexes are outperforming my portfolio
by a good little bit today.
It's a lot of that BMNR.
But we are heading into NVIDIA earnings after the close.
We also do have Snowflake that will be reporting,
and then we also have CrowdStrike
that will be reporting as well.
There's a few other names, HP, Viva, et cetera.
But on NVIDIA earnings earnings day thought and mindshare
is very tough to go around and the only ones that are getting it on a day like this are going to be
crowd shake and snowflake and that's even going to be a little bit all eyes are going to be on
nvidia again nvidia earnings 4 20 p.m eastern is the time the numbers generally come out and then i disagree what i disagree because
invidia's report to 420 and that gives crowd strike and snow 20 15 to 20 minutes first so i disagree
i know i think invidia is the important one but i think i think that they they will have their own
eyes on them for a while it won't just be about video I don't think that long I
think that it will wear off pretty quickly but yeah you're right no I think
that that's a good point in video going at 420 especially after MongoDB yesterday
last night no one's gonna be watching Viva and HP Q or net app or HP Q you're
watching because it might give insight into video how the service you're gonna
look at it tomorrow no I'm gonna looking at it tonight because I had nothing to do for
15 minutes between crowd
and video. I hate to tell you, HPQ
comes out at 4.15, missed opportunity.
Should have been like a 4.01.
It would have been a good day for that. I agree.
Go back to your thing.
No, you're good. You're good.
But I'm excited. Snowflake.
Snowflake's the one that I'm very
interested in. I'm curious to see where the story is at.
I was buying it in the $140s, $150s, $130 range.
It was cheap.
Now we're maybe closer to a point where, you know,
I want to see the story continuing.
I'm intrigued to hear what Snowflake has to say.
Mike, though, you can jump in a little bit.
What's standing out for you in this market?
I mean, Amazon's not looking so bad. It's's a little green but i guess underperforming the market
microsoft showing a little relative outperformance apple looks look decent out of the gate so i mean
where do you want to go here i mean spy just hit an all-time high about uh 45 minutes ago right
um this market remains strong uh what did i do today i didn't i did a couple things today i
bought amazon calls i'm long amazon calls i trimmed half my apple call position today into the highs
because it hit a trigger so i took it uh i traded some nvidia today off the lows because i thought
there was no way they were going to sell that off all day long I'm still sitting I sold my Tesla calls I bought yesterday afternoon to the clothes that
I held overnight and you know there's not a lot going on here there's nothing
bad Amazon looks good Google continues to look good you know Evan we'll see
what Nvidia says tonight again my gut instinct i don't i don't
want to go over this in detail is that nvidia beats and guides up is it enough here is my only
question and we've seen in video when they've done this over and over and over again for the last two
years sometimes it's just not enough depending on how much it's moved uh outside that it was kind
of a bounce back day you know you know um bitcoin had a nice bounce
back coin paid highs palantir and hood and meta got destroyed today for no particular reason but
it's just kind of how this market's working i i'm trying not to do a lot here i'm trying to keep it
simple and small and just keep moving around and shuffling my feet.
My long-term positions I haven't touched now in like three weeks.
Just trading in the day.
I'm just trying to find things there.
Apple looks good.
Amazon looks good.
AMD looks decent.
The financials still look fabulous.
Energy had a nice day, right?
We had a big bounce on energy today for no reason.
Kohl's surprised everybody with their earnings this morning. We talked about MongoDB briefly,
but I mean, that was an incredible report from them. Will it carry over into other sectors is
the question, like Snowflake, right? That's one to watch tonight. Overall, I'm pretty just neutral
here. For me, I'm working at half day at best on Friday. So I have a day and a half left in the market. We'll see what Nvidia brings tonight. I'll trade that and then I'll look to trade that tomorrow.
CrowdStrike brings opportunity. Snowflake, I'm still long in my long-term account, about 90% of what it used to be, but holding a little bit here into earnings. I'm not hedging it. I'm just letting it ride.
It's the last week of summer.
I know I keep saying this.
I sound like a broken record, but it's like there's not much going on here.
It's quiet overall.
Yep, that is fair.
It would be quiet without these NVIDIA earnings, and we'll see what it ends up looking tomorrow.
You're right. If we didn't have NVIDIA earnings and we'll see what it ends up looking tomorrow. You're right.
If we didn't have NVIDIA earnings this week...
We also have DCE on Friday as well.
So there will be a little interest on Friday.
But yeah, not a good time.
We also have Labor Day on Monday.
So we're going into a three-day weekend.
We have GDP tomorrow too.
That would be interesting.
Yeah.
I want to dig a little bit more into the NVIDIA earnings first here.
We got Monitive up here, and I know you dig into the space.
And then we can kind of see where we're at, throw the panel around.
And the good thing about NVIDIA reporting at 4.20 p.m. Eastern is we'll have plenty of time to talk heading into the numbers and dig into different parts of NVIDIA.
But Monitive, I want to give you the floor for a little bit and just kind of
generally talk what you're expecting here. I know we've talked a little bit throughout the days,
but maybe we have a new audience. I'm sure we have a lot of new people joining in as it is
a very big day. But what are we kind of watching here? The couple stuff that I've seen as people
have started to point towards content is China sales, that being a big question mark. And a lot of people have also pointing
to those CapEx charts
that I will pin one up in the nest above,
but it's basically looking at the hyperscalers,
the amount of money that they are spending on CapEx.
And it increased quite significantly
this past quarter from the two prior
where the two were kind of stagnant.
So not all that goes to NVIDIA,
but that seems to be one of the common areas
that I'm seeing people pointing to a lot
when they're talking about their excitement.
So kind of leading us into it, Monitiv,
I would love to hear just your thoughts
on these NVIDIA earnings, your take on those two,
and kind of what's going to be on watch.
Good afternoon.
So there's a few things that are,
I'm sorry, was somebody saying something?
No, I was just saying good afternoon, sorry. All right, no worries. So there's a few things that are, I'm sorry, was somebody saying something? No, I was just saying good afternoon, sorry.
All right, no worries.
So there's a few things that I'm going to be watching.
One, obviously, something that's become very important in the last few quarters since DeepSeek, that is margins.
So that's really a number that NVIDIA turns on.
So watch those margin numbers, net of one-time items.
Forget the one-time items for a second, right?
Just margins.
So that's the starting point.
Second is the product mix, which is, you know,
how those margins are derived.
There are plenty of reports out there that hop or sales are still pretty high
because there is limited availability
of the Blackwell series and there's a wait list there. So it is still for Nvidia,
you know, a shortage of their ability to supply or the limitation of their ability to supply.
Do we know roughly? So I saw a Wedbush comment and we don't need to go into how much it was, what it actually
is, but they said that for every one piece of supply, there is 10 demand.
Do you think it's that quite extreme or do you think it's like a lot closer, maybe like
two to one or, yeah.
I don't think, I honestly don't think it's that extreme, but there's probably two or
three to one at least at this point in time. Remember this, there's this continuing capex, but that large
number hardly is a replacement of hoppers too. So it's difficult to judge exactly how
much will be, you know, driven by a hopper replacement? Because, I mean, the earliest of the hoppers are probably, you know, three and a half, three years old now or longer.
So at the edge, there will be some, you know, there will be some level of replacement starting soon enough if it's not here already. So maybe next year, part of that will be, you know, some replacement as,
as Google, for example, or Microsoft, they still have a four year depreciation cadence,
they start writing off some of their, you know, some of their investments off, right,
fully depreciated. So I don't know, I think it's better than two or two or three to one,
So I don't know. I think it's better than two or three to one. We'll have to see.
We do know that there are multiple people that I trust reporting that, you know, there's a wait list that's longer than a year.
So that gives you some idea of, you know, how much could be.
you know how much could be and these are large numbers right remember you know we're talking
And these are large numbers, right?
about you know 50 60 billion plus in backlogs probably or more so so okay let's go back right
so so margin first mix which feeds the margin right hoppers obviously have lower margin profiles
certainly now compared to to to the. So, you know, that would
feed that overall margin number. And guidance, of course, always important to, you know,
see where the guidance is going. But the last, to me, the most important thing is, where
are we going to go with one-time items? We wrote off 5.4 billion and then for a brief while they were, you know,
okay to ship the products and then it got pulled back,
then it got okay, then it got pulled back.
So they have a bunch of these in inventory.
So if they were able to ship anything at all during the quarter,
you might see some reversal
of that write down.
But the problem is they also started manufacturing these again, you know, at some good pace.
Now they had to stop it again somewhere, you know, maybe a couple of weeks ago to three
weeks ago, they stopped the production again.
Those are necessarily going to involve more write-downs.
So do we have a write-back of the charge we took?
Or at the other end of the scale,
do we have more one-time write-downs?
So my guess is there's going to be more one-time write-downs.
So the size of that and the commentary around that
is going to, you know, know change the number at least for
this quarter but again you know the gut reaction is probably going to be based on that headline
number which you know if i if i'm able to quickly enough read through the details i might take a
position based on that if it's the one-time number that's bad and the overall numbers are still solid, I might add to
a position here. But that's the basics of what I'm looking for at a high level. Again, you have to
look at how much reserve capacity they have. These are things of the details that I look into their
to their numbers. How much capacity have they reserved for the coming quarter,
how much, you know, capacity commitments, basically, what is their CapEx like, and what is their commentary like, right?
So all those things, I think, put together will give you a picture of, to me, what is
the most important thing to watch out in the coming weeks is not that CapEx itself, CapEx guide is going to change for the industry,
but I'm more worried about the CapEx spend
versus the guide changing dramatically
far before the actual CapEx guide starts to go around.
So that is really an on and off switch companies can spend
when they want and stop when they don't
want. They don't really have to go out and say they're changing CapEx, they just don't spend it.
So I think that's where we will see the first sign of any issues if there are,
and that will probably occur well before we actually see CapEx guides start to get changed.
I don't expect any significant CapEx guide changes
for the upcoming quarter.
But again, I'll be looking at what they actually spent
versus what they said they would spend.
And that difference will tell us the story all by itself.
Thank you for that, Monitiv.
And just in general, if any of you guys have any questions or anything you want to throw out there,
if anyone down below has a question you want me to ask Monitive or someone else on the panel
as we're going through these NVIDIA earnings.
Again, also, we'll talk Snowflake and CrowdStrike.
Those are some decent sides, and there's a few other earnings as well.
This is probably going to be an earnings-focused space, though,
so we won't get too deep into technicals and, technicals and trading, except for Nvidia. I am curious. I'm also curious
in the implied move. We'll talk more about that. But there's any questions, the point I'm getting
at here, throw it in that purple five in the bottom right of your screen, whatever we want
to call that spaces chat, and we will ask some of those questions going throughout.
spaces chat and we will ask some of those questions going throughout um doc talk i want to
bring you into it early and just i know generally you're not watching earnings but nvidia is such
a big one around the market monotent kind of gave some insights there on on what he might be
watching do you have any any thoughts any just initial kind of maybe things you're questioned
you're excited looking forward to it, it's a big time.
You can't dodge it today.
No, I'm not planning on dodging it.
I obviously don't own NVIDIA, but
it's clearly a very important
stock. I mean, you know,
it's $4 trillion stock.
It's going to move the market.
Yeah, I mean, I expect it to probably have
an impact on semiconductor plays,
on AI-related plays where I don't have a ton of exposure to AI right now outside of Nebius.
I don't really have a ton of like single stock direct AI exposure unless you want to tie, I guess, Amazon and Tesla into that.
But in my mid in my mid cap picks, I don't really have much AI exposure.
So I guess outside of Nebius, I don't have too much to be directly worried about when it comes to sympathies to NVIDIA. But when it comes to the actual report,
I mean, I didn't hear all of Monitiv's comments. So at risk of being repetitive,
I'm sure people are concerned about China sales and are concerned about H20, the new H20 and how that's going to work and if they've already made some preliminary
sales, if there were other units that went to China in the meanwhile when H20s were restricted.
It also probably be worth looking at their other international markets and seeing if there's
bouts of strength in other areas. We've seen that in the past when there was China weakness,
we've seen Singapore strength, which you can connect the dots there. But it'd be interesting
to see if they have any strength in other international markets. But overall, it's just
important that they be, you know, considering the fact that the stock has done a lot in the past
three months. You know, it's come from the 140s to the 180s. Technically speaking, there's nothing really to,
there's no warning signs headed into this report.
I mean, the stocks.
Can I check in a second?
You said beat, but I, I mean,
I assume they're going to beat.
I think for me, it's all about the guidance.
Like how much did they raise by,
and is it enough to move the stock here?
Yeah, I think that's part of the beat.
Okay, sorry.
I just want to make sure it it clear for our users, right?
Beating is easy these days, right?
But do they raise and how big?
Yeah.
No, I agree.
The guidance matters a lot too.
Yeah.
And, you know, they've got to beat the streets expectations on forward guidance as well.
So, yeah, we'll see what happens.
I mean, to be honest, the price action's a bit quiet headed into this report like you know if you really wanted to look at the weekly and say okay you know from may 20 end of may 27th week
through um the first week of august it was like nothing but green right it was like what one two
three four five six seven eight nine ten eleven weeks in a row green for nvidia from end of may
through beginning of August.
So that's people buying it up into the report.
I don't know if that's going to skew expectations or not.
I haven't studied the expectations well enough from the street on NVIDIA,
but there were a lot of reiterations at Outperform this week for that name.
I believe there was actually a negative report out this morning.
Let me go check my notes here. Yeah, there was. It's not from a big shop it's from lynx equity they're a small specialty
equity shop but they they said uh the title report was we think nvidia is likely to fade we are
stepping away from the stock after years of being bullish and so they brought their price target down to 160 and basically just said, I mean, long story short, they said that a hackathon was hosted by OpenAI.
They realized during this event that GPT-5 runs on similar platforms to other LLMs.
They think the LLM space is becoming more competitive
and more commoditized over time they make allusions to XAI they make
allusions to the Gemini model and talk about how they're similar in
benchmarking performance and they basically are saying that Nvidia's
relationship with open AI is becoming more commoditized with other AI
companies and as a result the products are becoming more commoditized with other AI companies. And as a result, the products are becoming increasingly commoditized, so on and so on.
So that's Lynx equities argument, not my argument, just to be clear.
But they did have that note out this morning and they were negative on the stock.
But pretty much everyone outside of that note that I saw this morning was bullish this week
on the street.
And I'm just making sure there wasn't anyone else on the other side
um yeah pretty much everyone else is bullish this week yeah baird barclays stiefel uh jeffries
evercore jp morgan all raised their price targets into into. So yeah, the tier one shops are bullish
and there's a couple specialty shops
that are raising some concerns.
But yeah, it's gonna be an important report.
Appreciate the thoughts there.
I see a bunch of your questions.
I appreciate that.
I wanna keep going around, get some initial thoughts
and then I'll make some notes
and we'll double back in on some of the points.
Wolfie, you've been up here for a while,
and I'm sure you have some thoughts on these NVIDIA earnings.
If you want to throw some Snowflake or CrowdStrike quick thoughts in there as well, feel free.
But what are you watching on these earnings after the close?
Generally, the thoughts here have obviously focused around the China chips
and the forward guidance is what's going to drive it.
But yeah, any thoughts on the conversation you want to add?
We appreciate you being here.
Yeah, outside of that.
So the data center segment is going to be the focal point outside of that.
They had 39.1 billion last quarter, which is 73% year over year.
So the bar is pretty high.
So the bar's pretty high.
AI stocks are running hot.
There's a lot of like additional stocks that you can kind of look at to gauge.
You can look at like Oracle, Dell.
You know, people like to look at some SMCI.
I think there's too much noise there.
But yeah, so I think outside of the Blackwell ramp and the China stuff, like the data center growth is what you want to pay attention to.
So if you take a look at like some of the mega caps
and just kind of like mirror their performance
with expectations on data center and then the reality,
that's kind of like how they split the difference
on some of those stocks.
So, you know, that's kind of the main thing
I'd pay attention to.
I mean, it's a $4 trillion company, and then market's pricing in like a 5, depending on where you look, 5, 5.5% move.
So it's quote-unquote muted compared to, you know, some of the historical moves that it's had.
But the one-day event vol is sitting at 9.3%, and implied volatility is at 46%.
vol is sitting at 9.3% and implied volatility is at 46%. So it's pretty big from an options
perspective. Which leads me to believe that they could guide better, but if it was me,
I'd be in the camp of a similar quarter to what they had last time where it was like good enough they they hit on a lot of
things and then the people that are are gonna like pick at it will pick at it but you know that's
that's kind of like you know go ahead on that last point uh i was looking at a tom lee post today and
who knows whatever uh i'm a fan but could be wrong here. He was one thing, the chart that he was pointing to was going into earnings. NVIDIA has typically that month before been up 10% on average over the last three years what I was going to add to it is the stocks effectively doubled, right, going into this print off the low.
So, you know, with a setup like that, like they really have to show a re-ramp, in my opinion, you know, into the into the earnings for people to kind of get like really excited.
So it's not not effectively. It's more than doubled from the lows. Right. So, um,
could it go further? Yeah. But I, I feel like the, like the, the real performance on,
on their quarter could be something like, you know, some of these ancillary plays to look at,
like, again, like Oracle, Dell, um, you know tssi is one that i have a lab you know
uh uh what else some of these some of these energy names that are related to some of these uh natural
gas plays like i feel like that's where you know the the momentum move might might happen um barring
some sort of re-acceleration that no one's kind of like looked at. Just, I think 4.5 trillion or whatever it's sitting at, it's a big one.
So it's like, kind of reminds me of like that mid-phase where, you know,
when Apple first was the first of the trillion, two trillion club,
and it just kind of like installed it.
It's kind of like what I feel like here.
The setup on a technical basis looks fine, though.
There's other earnings, though, that I think we could uh highlight before go for it before we move on i want to say that honestly what i was
referencing honestly is kind of actually saying a little bit of the opposite i just pinned it up in
the nest above but he was kind of taught looking at the month heading into just the months so not
from the lows or whatever and in the last month and video is up 1.1 percent and he kind of was looking at here
versus on average over the last oh it was up more like 10 to 15 percent gotcha when you said kind
of saying hey it's relatively muted and flat heading into this and why maybe it could be
outperforming yeah this is necessarily going to be happening yeah but but i'm just saying that's
the stock is it was trading on the low obviously no one's going to get the exact lows. But effectively, it's trading 90 bucks on the lows. We're trading at 180. So it's doubled. And this is a $4 trillion company. It did have that washout in the last few days. I think it was like August 20th, 21st, somewhere around there, where it went back to basically the 50 day. And it's basically in a bull flag. So could it break out? Yeah. I'm not saying that it can't, I'm just saying that,
you know, for me, like as a trader, not someone who wants to own it long-term or whatever,
just, just this like momentary move outside of just,
if they were to get some sort of reacceleration that no one's, you know, foreseeing,
outside of that, to me, like the, the bigger move,
the bigger squeeze that you can get
on cheaper vol is going to be like the core weaves, you know, the other ancillary plays
that trade in tandem with it. That's kind of like how I'd view it. That doesn't mean that
this stock can't move. It's just, you know, at $4 trillion, like, no, that's a fair point.
That's kind of what I was getting at there. But the data that you provide above is very helpful.
It's useful to see that it's trading basically mildly compared to historical.
So the other one that I think that people really focused on outside of snow, because I haven't done snow yet.
By the way, I have a write-up on NVIDIA.
You can just look at my profiles there if you want to look at the exact numbers of stuff there.
But so you have CrowdStrike reporting as well.
So CrowdStrike, the expectation is 83 cents
and a revenue of 1.15 billion.
The interesting thing about CrowdStrike
is that it traded basically back down its 200-day.
There's been two other times that it's kind of had a similar move
in the last two years, or call it a year,
a year and a half, somewhere around there.
And that was in March of this year,
in August of last year, and then in April of this year, right?
So we've effectively bounced the last two times off the 200 day
that's kind of like where we're at right now you know I think a sell-off of what would it sell off
it sold off 20% into into into that 200 day basically if they could kind of hit and not
not do some of the things that some of these other cyber names have
done it kind of sets up for a possible mean reversion setup back to like that 450 460 level
back to that 50 day that'd be solid now if they were to come out and say something that you know
people weren't expecting let's say like that monotiv's got this theory um where he believes
at some point and i and i agree with him, so I'm not throwing it on him.
Don't want to sound like that. But at some point here, some of these companies might
transition from having best of breed to having an entire suite of products where they kind of
sacrifice from 100% best of breed to maybe say like an 80% good enough product.
So long as they get like an entire suite of products. And I don't know when that will happen,
but that's kind of like the, the thing that I want to pay attention to on, on, on, uh,
CrowdStrike. If there's any sort of, you know, uh, attrition or anything like that from customers
or, or customers are starting to bleed elsewhere. That's kind of what i want to pay attention to go ahead my retort on that would be
um if is crowd strike still the best in breed because that's what my kind of understanding
when people fix with it and then my retort if that answer is yes is wow if you're if you're a
ceo or whatever that cuts the cyber spend, either you guys are close to bankruptcy.
No, I'm not saying that you're going to cut the cybersecurity spend.
But go down to a worse option?
I would think that's when you would trade down less.
It's not about a worse option, you guys.
There's a big nuance here.
Nobody's talking about a worse option.
The point is when you talk to CIOs,
when you talk to technology leadership at customers,
one of their biggest pain points is having to deal with
10, 15, 20 different solutions
and integrate it all together.
And those interfaces are typically where the biggest,
you know, holes in cybersecurity are, where they get hacked.
And it's their responsibility for most
of those interfaces. It's their responsibility to maintain that because it's their code.
And it's very, very expensive and time consuming. And it's not even practical for most customers.
This is exactly what happened during .com or post.com too, customers like to consolidate to a fewer solutions. Now,
one of the ways they might do that is tell their different vendors to go work together and figure
out and have one person responsible for leading that solution implementation at their own site.
That's possible, but that necessarily means there's a give and take in margin between
those companies when that happens. So this is not about going to a worse solution. This is purely
about customers becoming more cost conscious and not willing to deal with 15, 20, 30 different
vendors. This is the reason why Microsoft still holds the leadership in the cybersecurity platform.
One, because they have the bundling power.
And I mean, they're freaking Microsoft.
So they pretty much own a lot of the software stack.
The largest software company in the entire world, literally the largest cybersecurity,
sorry, software company in the entire world is one of the best, I would even say probably
the second best endpoint client management platform across the entire world, is one of the best, I would even say probably the second best endpoint
client management platform across the entire board.
CrowdCheck, without a doubt, is number one.
But then now you have Palo Alto coming up, acquiring almost every single vertical in
cybersecurity, recently just acquired or announced the deal with acquiring CyberArk, which is
mostly an identity.
Nikesh Arora, the CEO, is literally
going for the entire platformization. That was sound and apparent last year. Nancy Pelosi went
long. Sorry, I had to slip that one in there because Pelosi never misses. But dude, that's a
giant. Also, then you have a lot of institutions who are like, well, I can't really invest in
CrowdStrike right now because they're not necessarily consistently profitable. If we do see that this quarter, we need to see net retention
rates stay above 120% to show that they are continuing to head toward a very fast-growing
cybersecurity company. But also, I'm interested to see how the CM, SEIM sector or segment plays
off of CrowdStrike. That's a triple-digit growth right there.
If that continues to be triple-digit, that might actually start to increase or reaccelerate
the revenue.
That being said, 20% haircut from CrowdStrike at their crazy valuation a couple of months
ago still brings us to an expensive valuation today.
So is the market pricing that in?
Or is the market taking it to be either a trader perspective?
Chart is on the 200-day, like Wolfie was saying, could relaunch again. Fundamental perspective,
this thing could just trade sideways for the next couple of years and not even be a fair value still.
So there's an argument there. But going back to our original point when it comes to cybersecurity,
you don't want to have a million vendors. You want to consolidate.
And like Manta was saying, consolidation can equal savings too, because then you have the
bundling negotiation as well. If you have like five, six, seven, eight modules from Palo Alto,
and I don't think they still count how many modules there are anymore. CrowdStrike only
counts like getting seven plus modules now. But if you start to head to that point where you're
starting to platformize
your entire cybersecurity horizontal
or your entire structure within your business,
then you could look for savings.
And if that's the case,
well, Palo Alto might offer a bit more
competitive solutions versus CrowdStrike
because CrowdStrike is not a cybersecurity
all-around platform.
Palo Alto is.
I'm not saying Palo Alto is any better.
I'm just saying
that that could be the direction they're headed yeah that's that's the point that that acquisition
and just having that entire ecosystem on the on the arr front um so in q1 their arr grew 22 percent
year over year which was a little bit below some of their analyst expectations.
So the net new ARR was 194 million, basically,
which was a sequential decline.
So that's, you know,
the trading at the premium valuation that it gets,
not having that suite of products like some of their competitors are starting to build out.
And then if they were to have any sort of slippage there,
that's kind of the thing that I would worry about
because that would kind of potentially echo some of the things sam and monitive was talking
about yeah and also well hey this is the fourth quarter reporting after the out is last year
right i don't think that that was something where it was like hey guys we need to sell
cyber security crotch right now i mean like obviously the company did have a bit of a sell-off
then traded to probably around fair value and then bounced right back to where we were today,
making new all-time highs, not today, but it did recently. They need to continue to execute.
That was what the market was betting on after that earnings last August. And we need to continue
seeing that today. I think profitability might come into question at some point. I'm not so sure
if it will
right now, but we need to see continued momentum. And that's what happened last quarter where you
had a bit of a slowdown and then the market's like, well, I don't know. Does it still deserve
a premium valuation for a company that's starting to decelerate its top line growth, even though
the ARR is still looking pretty good? But it's a growth stock. Is it still growing? That's really it.
You know, how's its earnings growth as well?
We flip into profitability the next couple of quarters,
like this thing could probably still maintain its leadership,
but we need to see it heading toward that direction.
Nice. I do appreciate that back and forth.
So it was good. I mean, that back and forth.
So it was good.
I mean, Wolfie's in it, so it's probably going to go up.
No, so I actually don't have a position like I used to.
So like 2022, 2025, the start of this year,
I had pretty sizable positions in both NVIDIA and CrowdStrike.
I've reduced them significantly.
I don't really have a position in crowdstrike i missed actually you know the last i don't know whatever that ramp was um and off the off the may to you know july run right so i missed most of that after after we
had that sell-off in in liberation liquidation, once I got a little bit above break-even,
I just, or not break-even, back to the breakdown point,
I just kind of cut it.
I thought there were better opportunities out there.
It's worked out for me, but, you know,
I don't really have that size position.
I have been looking at, you know, possibly adding,
but, you know, a lot of the things that Monitiv and Sam spoke to,
in addition to, like, sometimes these companies
get premium valuations, and then they might have the premium product, but someone might, might offer a premium suite. And then the
market kind of adjusts to wanting that, that then the premium valuation goes elsewhere because
people's expectations go elsewhere. But if I, if I, I don't know if you want to go through other
earnings things, or if you'd like me to just, I have a question and I, and, and I kind of want
to speak to something, But if you want to...
You can keep going.
Okay, cool.
So do you know what the best performing sector
in the stock market is in the last week?
Utilities or financials?
Energy.
Financials?
It's energy.
Energy is quietly...
The S&P 500 energy sector is up 3.86%
over the last week or so.
It's the best sector by over 1%.
I've asked this question probably like four or five different spaces.
No one's hit it.
Yeah, we got stumped.
Yeah, so, and that kind of, I think, speaks to how much things,
you know, how many different things are working in the market
many different things are working in the market and how much things can move. And then if something
and how much, you know, things can move.
is just unloved, people can avoid it and not even notice it. So I've been along quite a few energy
names. Valero is one I mentioned when I took it here a couple of weeks ago. Stock's up about 14%.
Options are up significantly. I trimmed some today, but i do think i've mentioned it a couple
times now if you take a look at some of these uh energy stocks uh some of them are breaking out of
you know year year and a half down trends some of the big ones that people pay attention to are
breaking down breaking out of year and a half down trends from april of last year um and then
if you take a look at even though like wti is like a terrible setup but if you take a look at, even though like WTI is like a terrible setup, but if you take a look at like positioning, it's quite skewed.
And it has it so that like large holders have the smallest position relative to open interest since 2010.
That includes when oil went negative in 2020.
That's a violent setup.
So that gives you a squeeze potential.
So I just want to put it on people's radar again. I mentioned it a couple of times. It's been
working. It's been working quietly under the table. But I do think there's a potential that
some of these recessionary camps or some of these energy glut camps that have kind of held things at bay for a while in tandem with some
of these inflation is cooling camps ideas. I think some of that stuff can unwind to some degree. I'm
not saying that we'll have rampant inflation or no recessionary caution whatsoever, but I do think
that a lot of it has been overstated to some degree. And I think
that if there's any sort of reversion in the opposite direction, that these types of names
could move very quickly. So if you just take a look at some of these stocks like APA, SLB,
C-Drill, EOG, CVX, XOM, Valero, et cetera,
like they're all breaking out, right?
And they're breaking out quietly.
So the moves haven't been made totally yet,
but I just want to put it back out there
because it's something that, you know,
I've been talking about
and I still don't see enough mentions of it publicly,
which kind of tells me
that there's still probably some room to go obviously these things can reset again given that the breakouts are still fresh
but i do think if they continue to base uh and the relative performance kind of stays there
any any sort of uptick in the underlying asset price of energy and those things can come back
and vote pretty quickly and they could rip so So I just wanted to bring that to your attention.
You know, one thing I will say, I'm not a big oil gas guy, but around the NVIDIA conversation
and the data centers and these power hungry, like these are extremely power hungry.
And maybe that's the advantage of NVIDIA because their chips are more efficient efficient but we have to find some way to be able to power all this stuff and even as they become
more efficient we're going to be continuously using more and more power i honestly i struggle
to get into some of these nuclear names at where they're trading at but i really do like want to
play this kind of energy theme that i'm kind of seeing but i don't know i struggle
because i have nothing against oil and gas yeah but it just doesn't get me excited no i feel you
i feel you if you are looking for a cheaper valuation play uh uh nuclear take a look at
floor floor owns i think 56 of sm, and they have their own core business.
Basically, you get all of their SMR investment basically free when trading at the current
valuation. So you can get a cheap way to get exposure without having to pay a high premium.
And they should sell some of their stake,
which would be accretive to their stock itself.
So I don't know when that will happen,
but that's a cheap one for you.
By the way, the ticker was FLR, Frank Laurie Rabbit.
Stock Talk, I know that's an area that you're interested intrigued in uh you've been
playing a lot of the nuclear plays um you got any thoughts on that kind of commentary for me like
like i know i want to play this this kind of energy area data centers and video we're talking
about today they're just going to use more and more um but i kind of struggle with these these
uh these nuclear names.
Yeah.
There's a lot of picks and shovels and leadership names you can buy.
You can own the power producers.
You can own Constellation Energy or Talon.
You can own picks and shovels plays.
If you really want revenue-producing plays, there's plenty.
Curtis Wright, ticker CW.
You look at Mirion Technologies, ticker MIR.
Marion Technologies does radiation safety detection globally.
They do it for nuclear medicine.
They do it for the nuclear energy industry as well.
Curtis Wright does your reactor coolant pumps for not only for nuclear.
Do we know for sure that nuclear is the answer i mean no one knows anything
for sure you know you invest in the things that you think are going to see traction and you know
you fade the things that you don't so i mean that's up to you that part's up to you you i don't
i'm not going to tell you what to have conviction in but if you have conviction in the electricity
thematic surrounding ai which it sounds like you do based
on your question, there's a lot of ways to play that. You can play that either directly through
electricity providers, power producers. You can play that through service providers and picks and
shovels to the AI industry. You can play that through data center power and cooling plays,
which also in many cases sell reactor coolant pumps to nuclear
power centers as well. You could find dual theme exposures that are overlapping if you want to go
into materials instead. Like you could go to a company like Materion, who has beryllium alloys,
which are used in aerospace and defense and nuclear energy and a lot of the hot thematics. So yeah, I mean, you don't have to buy a speculative
pre-revenue company to get exposure to these industries. We actually covered this on my last
workshop with our community. I talked about speculative plays when you're building baskets,
picks and shovels plays, and market leaders. And so there's plenty. Yeah. Power producers,
and market leaders.
And so there's plenty.
Yeah.
Power producers, infrastructure, radiation safety and detection,
reactor coolant pumps, CRDMs, which is control rod.
I forgot what the D stands for.
So control rod something mechanisms.
And, I mean, there's plenty.
There's plenty of stuff.
Yeah.
Yeah, Josh, do you join us up here?
Hey, yeah, yeah.
I figured I'd just step up just for a second
since you guys were talking about energy.
Wolfie, I thought you did a great job
sort of walking through that.
Stock talk.
The one thing I think about with this
is that a lot of the companies in the sectors
that you were discussing in subsectors,
even if they're not sort of pre-revenue, they trade at really, really high multiples on cash flow and even
revenue because of expected sort of future growth. And I think what Wolfie was saying,
and I think I agree with this, I'll sort of reiterate it, is that there's a set of natural
gas producers and services companies and sort of similar companies that
traded ultra low valuations because they're treated, they're treated as if they're sort
of this dying business when right now the U S demand is rising quite a bit, both for
exports as well as for these data centers.
And I went down the data center hole a little bit, just chatting with some friends that
are building them, developing them, looking at doing that.
And then some of the guys at some of the big tech companies, and apparently they think
they're actually way undersupplied and that there's going to be many years, at least on
the AI portion of the data centers, there's many years of development necessary, not even
for some of the speculative growth stuff people are saying, but just to sort of backfill for some of the existing solutions and existing plans.
And so I think the idea is just, hey, XYZ natural gas producer services company trades
at like three times EBITDA and like four times or five times free cash flow or, you know,
hire multiple on free cash flow, but growing 20 or 30 or something percent a year within cash flow.
And it's sort of a deep value way to sort of get exposure to this growth theme.
Everyone just treated it like it was going away.
And instead, these things are very valuable.
So I think that's sort of the thesis and I'm seeing it.
And it's pretty cool to get to buy some of these companies
that are already benefiting from this trend without having to pay any sort of, they're still like trading
as if they're going away.
Yeah, I appreciate this thought.
I did lead him that way.
So that was probably my leading question.
But yeah, that is some good points there.
Josh, I was sitting there during that call.
I can't lie.
I was like, oh, God damn, do I have to go in and buy oil or one of these oil names right now for another day, another conversation. I think it is quite clear
as we're going to talk about more on NVIDIA here that we're building more data centers.
These data centers take a lot of electricity. They're going to be getting more efficient,
but just we find more and more ways to need more energy, not less, and we need to get more ways to
do it. And yeah, there's certain scales and a lot of interesting conversations there.
Josh, I appreciate you coming up and hanging out with us, though.
Yeah, yeah.
And just one last thing on this, and this should segue for NVIDIA generally.
If you look at the bull case versus the bear case for NVIDIA,
they all involve tremendous amounts of more power demand
it's like the bear case implies that u.s power demand is going to grow by like three percent
a year for the next five years and the bull case is like it growing at like seven percent a year
so you're still either way it's an important conversation and what i was addressing was
more like the near term longer term it's possible nuclear wins and some of these high valuation stuff actually does quite well yeah i know i wasn't i wasn't being dismissive
of the nat gas thing evan just sort of teed me up and asked me about nuclear so i just provided
some names but yeah i agree there are nat gas names that are definitely undervalued versus
their nuclear peers but i mean i think market premiums shift as a function of sentiment, partially, and also due to like catalyst-rich environments that
have clearly been a lot of, you know, catalysts for the nuclear space. And there've been
a lot of direct policy initiatives that have raised investor sentiment around those and led
to multiple expansion. I don't know if we'll see that type of insane multiple expansion with the NatGas names.
Yeah, I don't know if I'm in that camp,
but I do think that there's definitely some value there for the value hunters
that are looking for attractive peer valuations for sure.
So yeah, no, I wasn't dismissing those.
I just responded to Evan's question.
Yeah, and again, maybe wrong tone. And then I wouldn't argue that you get insane multiples.
But I think if you look at the difference between where natural gas producers trade and where sort of general industrial companies trade, there's room for the multiples to essentially double just to be in line with sort of miscellaneous industrials.
So so you don't have to go crazy, I think, with these things to think that there's a lot of upside. Yeah, a lot of the charts look good, definitely. I mean, just from a technical
perspective, I mean, XLE's chart looks fantastic, actually, right here. Looks like it's headed back
to 95 if I had to bet on it. But I would like to see a little bit more accumulation volume,
as it comes in this 90 spot. But yeah, nice relative strength for XLE. All those things
are definitely true.
Been some nice performers in the index
in the past couple of weeks.
And I want to bring us back a little bit.
One of the things that Wolfie said
actually sparked a deep memory or something.
We used to talk a lot about sympathy plays
on this basis, like all the time.
And Wolfie brought up a good point that, listen,
NVIDIA is a $4.4 trillion company.
Whatever much it moves in either direction,
there are going to be underlying companies or other sectors that move more,
just given the size of where NVIDIA is and everything around it.
So I wonder if there are any sympathy names that you were watching around these NVIDIA earnings.
I wonder if there are any sympathy names that you were watching around these NVIDIA earnings.
It's easier to identify specific sympathies with smaller companies because you can get a more direct peer comparison.
Like, NVIDIA doesn't really have any direct peers.
And when you look at the ecosystem beneath them, like if you look at the smid cap ecosystem beneath them i would expect
more volatility in those days i think wolfie's correct about that but i don't think you need to
be specific i mean i think pretty much if nvidia if nvidia has a bad reaction to earnings all the
ai names and ai infrastructure names are going to trade down probably even the power producers
probably like everything related to it so i think you need to like hunt for a sympathy in NVIDIA's case.
I think with other stocks,
you probably want to be a little bit more specific when you're looking at
sympathies.
But with NVIDIA,
I don't think you need to be that specific.
I think anything related to the AI trade trades down if NVIDIA trades down
tomorrow.
So I agree,
but I think you can,
you can take a look at like core weave,
for example, which has been beaten up. Take a look at like core weave for example which has been beaten
up take a look at dell dell has earnings next week so you know it's like one of those if it
if it doesn't um if it doesn't disappoint that's one that they'll probably run um another one that
has earnings next week is avgo broadcomcom. So, you know, last time around
when NVIDIA reported earnings
and we had that giant space,
that was kind of like, you know, post-print
while we were still on the call,
that was kind of like my pick.
It was like, hey, if you like these numbers,
I think you could buy Avgo.
I think that's the run-up into earnings.
It's exactly what happened.
They ran the stock after NVIDIA's earnings.
You kind of had like a four-day run
on Avgo into earnings
and then post the print, they basically sold it back to its 20 day. Similar setup this time around.
I think, you know, if you've got a good, you've got a good print, the ones that have yet to report
that are, you know, pretty much trading the same type of setup, type of name those are the ones that would probably
probably be the quote-unquote no-brainer trades moving into their prints next week
oracle probably two moves with it yeah oracle oracle 100 but oracle the the one reason i
didn't mention oracle is they had a headline i think it was last week where they kind of basically
guided guided down something uh some of their some of something, some of their data center stuff.
Yeah.
So I don't remember the exact headline, so I don't want to misquote it, but I do remember
reading a headline last week and that was basically the catalyst that kind of sent it
back to its 50-day.
So from an Oracle perspective, that's why I avoided it.
Microsoft got rich, so I want to avoid that one.
That one was pretty linear as well.
But yeah, I know Dell has earnings and Avgo has earnings.
And CoreWeave basically is going to trade in tandem.
And on the back of CoreWeave, CoreWeave has a 50% retrace back to that $90 level off the all-time highs.
It's held it now for about four days.
It's been battling it.
Today will be the first close above its five-day
since it broke down in early August.
So, you know, any kind of follow-through, any kind of that.
I saw some call buying as well,
so I don't really rely on that stuff that much.
But any kind of follow-through,
you basically got to set up for a possible 10 15 upside um before it hits you know your next cluster of resistance yeah i think all
the to by extension that all the data center plays really are are relevant there right like even like
look at names like applied digital iron even the smaller ones right like anything that's
by extension related to data
centers is going to trade in sympathy with nvidia and and for people in the audience we're not saying
these plays are definitely going to go down it's just that they'll likely trade in sympathy to
nvidia uh wherever it goes i don't know if it's going to go up or down but sorry sam you want to
jump in go eight percent of the market so we're definitely going to see like sympathy play across
the board for the most part i mean
it's gonna move the market not because the fact that it's eight percent it's because these other
sympathy plays are mentioning like we're probably talking like 30 of the entire market is gonna
trade in sympathy to nvidia which is definitely a large weighted sector when you think about it
um i suppose you even mentioned nebius man Nebius will likely catch some wins if NVIDIA does have statements about data center expansion. That's going to be wholly bullish for Nebius as well. you know overall if you continue to take that theme out there that look we we are still seeing
demand in terms of ai workloads that's just going to be tailwinds for a lot of companies out there
that have to do with non-nvidia plays for me personally nvidia was something that sold at 140
earlier earlier this year and i haven't bought it back since then uh not for a long term holding the
reason why is because well one it was already like a two three trillion dollar plus company at that
point so you're not going to get a crazy hundred ten times anymore but also i mean how much more
room can this growth actually grow grow in terms of eps growth that's where we saw a lot of the
appreciation nvidia uh in the last couple years at this point we're not going to see that kind of growth anymore,
or at least it's going to start to stabilize.
But when you think about these other companies, like we were mentioning,
we had Nebius, of course.
Corvi was a very strong momentum play as well.
You had Astera Labs.
I mean, even though it's not a profitable company,
so it might not be in a lot of sectors per se,
they are still a very important company when it comes to NVIDIA.
So along with these other data center plays, if, if Jensen Wong says like,
Hey, we're looking into, um, he's not going to say it directly, but if you hear news that
they're going to be switching out to whatever, switching out to whatever supplier, where's
that coming from?
Evan, Evan.
Um, okay. So if, if we, if we hear any news yet yet that's going to move a lot of data center place
as well but i think also think china commentary is going to be pretty important monitor was talking
specifically about the numbers but i think that commentary is going to be pretty important there
um there are hyperscalers in china that do need these NVIDIA chips against whatever argument we have here.
You know, you do need the China does need the leverage in terms of tariffs where like
they don't want to be like, hey, we really like these NVIDIA chips because then they're
going to lose a lot of leverage in terms of that negotiation.
But if we do hear some positive effects of that, you know, that might mean good for data
center output.
Right.
So if China needs a lot more H20 chips or Jensen says that there is still heightened demand for these chips and they're going to restart a lot of the production, you're going to see a lot of sympathy plays like a stare aloud.
So like I said, it's a major component with a lot of these.
You're going to see Micron likely catch a bit off of that one because they do provide the HBM memory or high bandwidth memory for a lot of these chips and stacks as well.
Avgo or Broadcom's Fogner rally off of that because they have to do with a lot of network throughput as well as well. Avgo or Broadcoms, Fogner and Rally, off of that because they have to do
with a lot of network throughput as well.
Marvell technology and so on.
These other dependents sympathy plays
will continue to rally
even if NVIDIA is enough like 5%
because they don't need a huge market share
in order to rally.
They just need some,
especially in these sectors.
I like the logic there,
but I think they will still trade with NVIDIA.
Yeah, I mean, NVIDIA's down like 5%.
I doubt.
Yeah, they're not going to be able to do it.
I have one for you, Evan.
If NVIDIA doesn't run or it's not as ideal as everybody wants, just buy Apple.
They have no AI exposure.
Wow, man.
You need something for them? I i love it what a guy you looking
for something like what do you want i bought i bought apple i told him yesterday i bought apple
i'm up pretty pretty nicely you know current going into tonight or going into but buy the hype into
the iphone event yeah i like it it's not a bad idea but um you know i i do think that if if they
start selling off the AI names,
they got to hold shit together somehow.
They'll probably just bid Apple.
People will just be like, no, this piece of crap, why is it up?
I mean, if you want to dodge a bigger speculative correction in the markets,
NVIDIA doesn't have much room to fuck around on these earnings,
if we're being realistic, just from a price standpoint.
It really can't go below 172 yeah you sell below
172 which is nine bucks below where we are right now that's not an enormous move then yeah it'll
that weekly structure starts breaking down and that will be brutal for probably for the whole
market not just for nvidia also kind of suss that sP 500 is making all-time highs and NASDAQ hasn't.
You know, that's like, it doesn't really look too good in my opinion.
I mean, I don't know.
I mean, there's a little bit of hesitancy involved.
It's not sus when you look under the covers, man.
Yeah, I have to say, it looks like NVIDIA is coming into this one price for perfection.
I do agree with Stock Talk there.
Yeah, I mean, if they are going to miss, they can't.
I mean, the stock can't afford to go down too much.
Like, if the stock goes down a couple of bucks on a miss, that's okay.
But if the stock goes down 20 bucks on a miss, the chart is going to break down.
And you could probably see 140s again maybe 150s so
yeah just have to be cognizant of that can i ask you something though yeah a fair failed
bearish breakdown what is that yeah that's okay i mean if the stock let's say the stock sells
let's say this happens let's say the stock sell like has's say this happens. Let's say the stock sell, like has a negative reaction,
sells into 172 and then bounces and goes green.
That's incredibly bullish in the market.
We'll probably get a bit off of that,
but that's like a pitch perfect scenario.
I mean, alternatively, we just rip from here, right?
I mean, you have 185,
which is the top of the local range overhead
from two weeks ago, 184.50 really.
But yeah, alternative scenario is you just rip through that 184.50 really.
Alternative scenarios, you just rip through that 184.50 and make new highs. That's
also possible.
I'm just saying there's not much room
for it to dump
is the point I'm making.
We've got about one minute
to the close. Snowflake and CrowdStrike
are around 4.05,
whatever, around there nvidia puts
their numbers out around 4 20 p.m eastern so we'll have a little bit of time still but the market is
about to close monitor your hand up yeah uh just just a couple of quick points to add to sam's
point yeah china says one thing and does a whole different thing behind the scenes so it's very
possible that you know while they're signaling their large customers to not buy H20s, they don't have an answer in the meanwhile.
And they are not, for them, it's a bigger loss of face to, you know, to fall further behind in the AI race than to, you know, quietly end up, you know, accumulating H20s without, you know, while signaling something else.
So they need it.
It's not like they have an answer today.
They have a bigger desperation to get an answer faster, but that's still, you know,
year or two or three away, you know, at best.
So, so there isn't much of a choice they have right now.
The other thing I want to add on the list of companies you're talking about is
sympathy moves, Credo has not yet reported. So that reports next week. That is in the network
retimer space. They do have a great business that's growing similar to Estera, but more in
the network retimer side of it. So that's another one that's very interesting.
very interesting.
That is an interesting one, actually. Thanks for mentioning that.
That is an interesting one, actually. Thanks for mentioning that.
I also appreciate you having
the bell in the background.
What did you say?
4.20pm Eastern for NVIDIA.
CrowdStrike and Snowflake should be at
4.05pm Eastern.
As you heard with Monitibs' bell, the market
did just close there.
I at least have Snowflake up on my
screen if one of you guys want to get CrowdStrike
so you can see when that goes moving.
Just around earnings,
sometimes it can be difficult.
Sometimes companies will put on a sheet
and say it's on our website
and it takes two, three minutes.
Sometimes there's just a whole filing
and you have to go through and dig through it.
So it takes a second, two after they come out.
So stick with us.
And then also sometimes in this time,
there's weird, like a lot of happening
for two, three minutes,
and then nothing for two, three minutes as we wait.
So yeah, this is kind of just the more
live earnings part of the conversation.
But NVIDIA, 4.20 p.m. Eastern.
We have a couple names at 4.05 to get through,
and then we'll kind of shift more really in on nvidia 100 there's a
more than a few names net app is reporting at well net app is already down newton x is down 7 net app
is down four percent bill is up three percent that's really interesting you have urban outsiders
as well dude so tomorrow let's bring up American Eagle cuz what a second deal
they closed today with Travis Kelsey smart marketing team honestly I didn't
even see that they yeah did a deal with Travis Kelsey a little partnership or
whatever pulling up on the Sydney Sweeney why they made him a creative
director I thought it was something I don't. All I know is they're working.
There was some sort of thing there,
and I was like, smart headline.
I believe that's correct, Wolfie.
Yeah.
Who the thought?
Yo, Evan, could I talk some numbers
on the ones coming out at 405?
Can you give me Snowflake and CrowdStrike?
Yep.
Snowflake, we're looking for an $18.70 move or 9.34%, 11% up since the last report.
On CrowdStrike, we're looking for a $28.16 move or 6.62, and we are down 13% since the last report.
Nothing really crazy on CrowdStrike's previous reactions.
Snowflake, three quarters ago we saw plus 32.71%.
But that's pretty much all that's exciting
about those two the third one that we all know is going to be some crazy numbers to present to you
guys yeah i'm going to give you a second to go through that after nvidia will be 420 pm eastern
five below just reported to be that they had a prelim of 81 cents for 62
They had a prelim of 81 cents for 62, 1.03 billion versus 995 million consensus.
Take a look at that chart.
$50 on the lows basically this year.
Did you say it was up or down?
It's up.
First blush is up.
Full year, 4.76 to 5.16 versus 4.73 expected.
Sees revenues at 4.44 billion to 4.52 billion versus 442 billion expectations.
So they beat across the board.
I do see a lot of people joining in, so I apologize.
While we're talking about other stuff, I will keep talking about this.
NVIDIA earnings 420 p.m. Eastern, a couple names come up before.
So if you're just joining in, you haven't missed the numbers out.
And there's honestly going to be still another 10 to 15 minutes after this,
after 405 of digging in specifically to that.
Bic, you can keep reading some of them.
I know there was a couple names.
Again, I'm just reading out those two,
those CrowdStrike and Snowflake because I think they're the big dogs.
HPQ, I know a lot of people watching a lot of software names,
but those are just names we will dig into a little bit tomorrow. Snowflake because I think they're the big dogs. HPQ, I know a lot of people watching a lot of software names, but those are just names we will dig into a little bit tomorrow.
Snowflake still has not reported.
We're about a minute away.
Snowflake can be a big mover sometimes.
What's the implied move for Snowflake?
I don't know if Sniper you have around what that was
or like what the historical move.
Yeah, I got you.
For Snowflake, your implied move is $18.70 or 9.34%.
I feel like the implied move is...
I mean, it wouldn't be far off that.
The implied move?
Yeah, it's...
Sorry, the historic move, the historic move.
That's a Wolfie question.
The question is implied.
Impl implied versus historical
right now on the options for on 57%
for Snowflake.
Okay, so what that's telling you
is that compared to previous
it's 50% smaller.
It's 57%
of what it's historic.
Snowflake's initial move is up.
Snowflake is moving higher.
CrowdStrike as well.
I'm seeing...
Yeah, Snowflake's initial move is up.
CrowdStrike's initial move is...
Honestly, I'm not seeing a move yet.
Maybe I just need to refresh my page.
So Snowflake's revenue revenue is 1.09 billion
gross profit 788 point 2 million EPS minus 89 cents crop strikes not moving much yet half a
percent up in after hours so from what I am seeing and I'll let the numbers come in that revenue number for snowflake is a beat let's see
i'm sorry i got one stock up nine one point one four five billion on snowflakes revenue and 35
cents in their eps snow up nine nine nine and a half percent
then there we go snowflake moving higher there wow dumper and then all right on some of these
other names yeah I'm still give me a second I want to be able to understand the earnings a
little bit better well we'll actually move on for now. But
Snowflake, CrowdStrike, those numbers are out. Snowflake moving higher. I'll take a second,
look through and kind of be able to present it to you. Let's dig into Nvidia earnings though.
4.20 PM Eastern, Nvidia is reporting. That's why I'm sure pretty much all of you guys are here.
I want to circle it back to Monitive a little bit and just kind of, if I could, I know we started the spaces here,
but maybe you can give us a one or two minute brief of just kind of what you're looking through on these NVIDIA earnings
and the one or two parts you think that are going to end up moving us here.
Or at least, you know, maybe that's a little bit of a leading leading way, but the part that market watching what's going to move the stock.
Like I said, the simplest is going to be the guide, right?
It's as long as the guide is strong, you know, much of the other little things will be forgotten.
Or if the guide is really strong, you know, nothing else will matter.
Or if the guide is weak and the numbers are strong, it'll still be a problem.
Right.
So, so, so? So start there.
But again, it's the details on the numbers that matter a lot more to me than the numbers themselves.
So it's going to move a lot on headline numbers.
But I do expect one-time items.
So I would, you know, dig deeper into the numbers before, you know, I jump to a conclusion
and market always jumps to a conclusion right away. So so there will be significant one time
numbers in my opinion. So I will I will wait and adjust for that. Again, I want to read into the
details. Remember, these guys month closed at, you know, 27th of July. So so you know it's a month off the others. There
are some you know other things I forgot to mention for example quarter over
quarter TSM production numbers are actually down year over year they're up
22 and a half percent but quarter over quarter compared to last quarter they
are actually if I take NVID Nvidia's quarter as an example,
they're down a little bit.
The thing you have to adjust for,
which makes it very difficult is Nvidia makes up roughly
17% of that number for TSM and Apple is still,
you know, 21% and there's a few others, right?
So I don't know if who is driving that, if that's a read into NVIDIA, but it's worthwhile to
pay attention to that.
So that's from TSM's manufacturing numbers they put out every month.
So that's a bit of a concern, if nothing else, to watch for.
Again, like I said, one time, again, additional one-time charges against China
to see if there's any China revenue this quarter. You know, if anything was pushed through the
supply chain into Chinese customers' hands, that's important. And then the margins, right?
And the rest of the numbers follow from there.
those rest of the numbers follow from there
yep monotiv I appreciate you there so you can go through and give us a couple
of numbers really quickly and we'll keep going regarding in video yeah implied
move ten dollars nine cents five point fifty five percent previous reactions 0.55%. Previous reactions, plus 3.24%, minus 8.48%, plus 0.53%, minus 6.38%.
Change since last report, 34.83%. Open interest, new high, 19,119,605.
That is the numbers that I have for you on the end of the day.
Ooh, interesting. What's up, Empire? We appreciate you being here. That is the numbers that I have.
Ooh, interesting.
What's up, and by the way, we appreciate you being here.
They're buying Onam out for supercharging their CM platform.
This is that triple-digit sector I was talking about earlier,
but they're still down 5%. Interesting.
Interesting, interesting.
Quick question.
Yeah, Malatav?
If we consider the implied move to actually happen
in nvidia would that make it the largest one day move in a stock at the size yeah probably i mean
that'd be a 250 billion dollar move uh roughly it would be a 400 billion move
10 yeah i thought you were saying...
He said 5.5% is the implied move.
It's a $10 move, Monarch.
Oh, okay, got it.
Sorry.
Wait, you have $10?
You said NVIDIA has an implied $10 move or 5.5%.
Oh, yes, yes, yes.
Sorry, sorry.
My bad.
You are good. All right, we're about nine minutes away from these numbers going out did you read what we're really watching here what yeah did
you read crowd strike uh yeah we read all that out sorry so what we are watching here though
kind of summarizing that china is is the big question mark. A lot of the arguments that I've seen so far
have been on the moving on from China,
just in general,
is on the upside for NVIDIA
has been the CapEx spend.
When you look at the chart of the CapEx spend
for all these hyperscalers,
it was around $80 billion,
what was promised this past quarter.
That is up a good bit
from where it was at the past two quarters.
A lot of people have been kind of pointing towards that.
It is not a one-for-one going to NVIDIA.
There's a lot of other companies that kind of has an impact on,
but they're definitely a good piece of it.
And Monty did make the good point earlier that all of these annual CapEx spends are,
you know, it's not, they can, they, their promises until they actually go in and do
it. And we'll see when it's time. And if the market continues like it has, they'll probably
spend it and more. And if there's a pullback, maybe, maybe it does pull back quick. I don't
think today's the start of that, but we shall see. But yeah, so that, that's kind of the upside.
I think part of that also upside is during the quarter, they took a very big write down.
It feels like a very long time ago, but it was a $5.5 billion write down from the H20 chips that they were planning to sell to China.
And then at some point, whatever it was, they got a week, they got two weeks, whatever it was, to be able to sell chips there.
We got no information on how much went through.
I don't even, can they still sell? There was a thing with giving 15% of the sales to the US
government. I don't know if that ever happened. I guess that's a conversation to dig back into
for another day. But there are some question marks around that China segment here. And
one thing I will say is NVIDIA did the way they're kind of guiding is that they took away
all of, and Manitib, you can kind of confirm this, they took away all of the upside from China
in their forward guidance in previous quarters. So anything that they would have gotten would be
just incrementally positive on top of it. Is that correct? Yeah, but they did have a little bit of reminder left to take charge on.
I think they said the cost would be 6.9 billion or 6.6 billion if-
And they took the $5.5 billion right down.
And they took 5.4 against that. So there's still about,
which is why I say that plus, you know, the production started again and stopped again,
there is a cost, right?
Coase capacity, you have to pay TSM, they don't care about this.
There is a certain amount of throwaway, right?
You can only recover part of the hardware, the rest of the components are unrecoverable,
mainly memory, which is a significant part of the cost of a GPU.
So there is a write-off possible against half-finished inventory that they again restarted a few
weeks ago and then stopped again three weeks ago.
So that one-time charge could be another five, six billion, possibly.
I don't know.
I'm just saying that is a risk.
Yeah.
So 1.5 was still left
and there are some fixed ones.
But there are question marks around it
and there haven't been a lot of question marks around
NVIDIA over these last numbers.
When's the last
time NVID your missed earnings because
i can't remember that i was looking and at least on revenue it looks like 2023 there was one or
two number oh that's eps there was pre-tend dpt yeah eps though i think there was one or two misses
actually at one point so yeah they're not gonna miss clearly
the other video videos reported at 10 consecutive quarters with both upside
EPS and revenue I wonder if they're gonna do it again I mean I feel like
they're gonna do it again but that the problem is that is now the expectation
and if I'm correct and we are previously pre-released Anix 1 two years ago.
If you remember that quarter oven, so did AMD a week before.
And that was also due to some channel stuff.
they pre-announced the write down
They pre-announced the write-down.
yeah so the question is
how big the beat will be
not if you know they will beat
and to be honest everyone I know
is just looking at the guidance
Gorg the real trouble
this time versus last time
was last time was a clear cut policy change
that they said they cannot sell
this time around it's changing on a daily basis what the policy is Last time was a clear-cut policy change that they said they cannot sell.
This time around, it's changing on a daily basis what the policy is or is not, right?
You can sell, but we'll tax you, or we won't buy it, but we have demand.
So I don't know if they know enough to even take a charge.
Maybe that's why they've not pre-announced.
Evan, do you have the numbers for the next quarter for Q4 and Q3 no what is what is the forward guidance expected uh so around 46 billion is you know what the market is looking for that's
for this quarter that's revenue this quarter uh oh okay so Q3 so okay Q3 october is 52.1 billion on bloomberg okay interesting the full year is
199.7 so 200 million dollars and next year 250 billion damn up to 52.1 is that seems like quite
a big growth what's the eps expected the eps is just above $1. $1.01. 49% here over here.
Is that the same thing you have?
That's forward guidance?
No, that's the current one.
I have $0.98, but
yours is probably going to be the one that gets printed.
$1.01.
And do you know what forward guidance
for EPS was?
I don't see that currently.
I only see
8 people, so 112.
Can I ask,
what does it say for data center revenue?
That's going to be a really important segment. Last quarter
was 39.1 billion.
I think this one was 41, 42
is where the number will be about.
And that's a
very watched segment as well.
Let me look.
Data center.
And do you want for Q2?
No, this is Q2.
Okay.
Data center consensus is 41.3.
That's for this quarter.
Correct.
And then what about next? It ranges from
$39.7 billion
to $47.5 billion.
It's like,
you know,
very, really
big range.
What was the high end?
Sorry.
Higher end is
$47.5 billion.
I just heard the
$39.7 billion.
I was thinking
that guy's dumb.
Yeah, it's almost like...
It's probably
a wonderful person.
Don't tell me who it is i don't want to know
they're a great person nvidia two minutes away though are you ready i'm excited i am excited
does anyone have any predictions that they want to make about the move i'll come out and say it
and for the record i have some either some etfs pinned up in the nest above. I know we were talking about sympathy plays,
but maybe those move as well. Or maybe by
the inverse ones. I say
we up 5%. Let's do it.
Oh, really? I'm saying the options
buyers win it. You guys,
the law of big numbers, it should have
come in a long time ago. Also,
guys, watch it for a buyback.
Last year, in
this quarter, we had a $50 billion buyback,
so maybe we get something again, too.
All right.
It should be coming out any second now, the NVIDIA earnings.
The earnings call will be at 5 p.m. Eastern.
We'll listen to it in here.
Definitely stick with us.
But numbers should be out in a minute.
Sometimes it takes half a second, but you'll see the move instantly.
So you can get your NVIDIA charts up.
This 0.5% up before isn't really anything.
All right.
No one else joined on the other.
No one was big enough to say.
It stays within the market maker move.
It stays within the market maker move.
I'm with 20 seconds to go.
No, you can't get the whole move.
You got to take half of it.
The higher end or the lower end of the range? I'm gonna say lower end.
I think it's gonna be down 3%.
Last three have been gap up and they gave it
all back the next day.
Shut up. That's so
mean. Why would you... Don't put that on
me, Ricky, Bobby. This guy joined in and just says
something mean. Alright, we
are at 4.20pm
Eastern.
Don't do this to me, NVIDIA.
Don't make this whole thing.
Yeah, Magic Fairy.
Magic, we don't report for like 20 minutes.
That has happened before.
NVIDIA has done that to us.
Are you seeing the numbers, Gurg?
Stocks up a 0.6.
No, nothing.
Nice.
The earnings call confirmed 5 p.m. Eastern, so they have to.
It's moving.
I don't know what's going on.
It is?
It's moving.
It's up a little bit.
Nah, that's not a move.
Numbers aren't out, but futures just four trillion company it is a move
now it's down nvidia is out a dollar five dollar one was the estimate
um the number revenue numbers doesn't seem correct.
44.1? Let me double check.
Yeah, that would be a miss.
46.17 on Bloomberg.
Okay, and that would be a beat.
The double beat, at least on the numbers.
It's kind of in line.
EPS, I'm looking at
$1.05 UPS.
$2354 is what I'm seeing.
$46.05.
Look at the E-videos.
That's the T-500.
I'll take a look in a second lost margin of 72.7 data center revenue where are we revenue expected to be 54 billion lost my most yeah you said
what was expected 52.1 so forward guidance above Evan you got a typo in your post
What I said you said missing I see it
60 billion share buyback
60 billion share buyback
nice okay i would write about that okay
all right i got it out uh data center was a mess yeah data center 41.1 billion 56 percent year
over year five percent quarter over quarter that a miss. Blackwell data center revenue, 17% growth in terms of revenue for that.
72.7% gross margins non-gap.
$1.05 beat estimates, 46.74 billion beat estimates, slightly 56% growth.
slightly 56% growth.
Q3 guidance, $54 billion.
Q3 guidance, $54 billion.
Acceleration, a reacceleration
of 59% year-over-year for revenue.
Estimate was $45.9 billion.
Gross margins for next quarter, 73.5%.
Expectations for forward guidance
was $52.1 billion around there.
Oh, okay.
Maybe 45.
Yeah, no, my bad.
That was probably for guidance
from earnings up or something.
Okay.
But still beat, still beat. They don't have earnings. Whoa 45. Yeah, no, my bad. That was probably for guidance from earnings up or something. Okay. But still beat.
Still beat.
They don't have
earnings.
Whoa.
Don't throw them
under the bus.
What was the
data center?
have forward guidance.
They don't even
That was a few
hundred million.
You can stop.
Just a few hundred
million.
It's like 41.2%.
2.2%.
So what were the
numbers on data
center?
41.4, market wanted
41.29, so
rounding, yeah.
But they
mounted, correct me if I'm wrong,
they usually guide up, what, 4 to 5 billion
more than what the analyst expectations are?
Yeah.
So it's below, so it's slower.
Right now, they guided
54, and on Bloomberg, the consensus...
52 is what I saw.
The consensus was...
Let me double check.
You said 52.1 before.
Yeah, 52.1.
So slightly below.
But it's, you know, almost on the dot because they always give up and down a billion with 2% margin of error.
Did they say they sold no H20 chips to China?
Yes.
Yeah, that's supposed to be quarter two.
Benefit from the...
Wait, so there were no H20 sales to China-based customers in the second quarter.
NVIDIA benefited $180 million release of previously reserved H20 inventory,
approximately $650 million in unrestricted 820 sales to a customer outside of China.
So they redirected some of the sales that would have gone to China outside of it.
It's coming right down to that 172 spot we were talking about,
so let's see if it can defend it here.
Well, they just put a low bar in.
Look, they put a low bar in, right?
So their guidance, they're saying H20s,
they're not really even backing that in to their guidance,
which is a smart move.
And if they do kind of ramp those back up,
they'll kind of be icing on the cake a little bit, right?
So sitting low bar i think
but if i'm correct even last quarter they had the same exact thing the guidance did not include
the channel numbers correct correct well yes correct and they're carrying that over right
they're carrying that over just assuming that they're not going to have each 20 sales for the quarter.
So this is the first time in 13 quarters when the beat was this small or not significant.
Hey, Evan, you know what they're not selling?
What are they not selling Apple
boom great call and listen NVIDIA is holding in so we're still within the range I'm seeing down
3% as stock talk was saying earlier the range isn't that big where we fall off so you know
I mean this isn't the when you look at the, I'm sure Stock Talk is still around that level.
Yeah, if it defends 172, it's gravy.
And it's doing that currently, so...
I'm still waiting for the books, right?
They've not posted that yet.
I don't see a cash flow statement.
I don't see income statement in that one.
It's already bouncing pretty aggressively here off the after-hours lows.
It's down just 2.5% now.
If it fell below that 172 spot, I would have said that's concerning.
You still expect non-gap margins at 70% by the end of the year.
Still good, still good.
I'm trying to calculate gaming right now.
What were they after this quarter?
Where did they say 70%? It says mid-70s.
Yeah, mid-70s is what?
Gaming was $4.3 billion on Bloomberg.
Market wanted $3.82.
Oh, there you go.
You got the acceleration beat.
Also, networking came in above estimates.
Networking came in at $7.25.
Market wanted $5.07. So $ 7.25. The market wanted 5.07.
So $2 billion there.
A billion in gaming.
So here's, wait, can I ask you something quickly?
So I want to look, this forward outlook thing is pretty big here.
We were talking about that for a second.
NVIDIA said next quarter, $54 billion is what they expect.
Wall Street wanted $52.1.
The precedent NVIDIA has set is that generally
they would have brought in like maybe let's say 55 billion but the kicker is here like last quarter
they assume nothing about china shipments in the quarter when was all of this china debacle stuff
going on with with these h20 ships there's another quarter there's a further piece here that says
they've not assumed any h20 shipments to China in the Outlook either.
Yeah, yeah, yeah. That's what I was saying.
Going forward, they're not counting on it.
What we were talking about, yeah, they are sandbagging the guidance a little bit, or at least maybe they always do, but they're giving themselves more upside.
Are they sandbagging more than normal?
No, they're not sandbagging more than normal no they're not saying begging more than normal and
you also have to understand this street analyst estimates those consensus some analysts have
backed in those age 20 sales because they do have the ability to technically sell them now
right so can i tell you something though should they not be sandbagging more now because the
uncertainty around these china chip supplies over the last
two quarters is a lot more that's a meaningful part of the business they were told they were
told they can sell them with their license and pay 20 now right i have no idea so when did all
this happen have they sold anything no this other than q1 no but I'm saying this whole last one that we were having...
The last part, I think, was outside their quarter
because their quarter closes
July 27th.
That's my question. Did we get anything?
And I still think they would say
they assume nothing if they put out
$500 million of chips.
Is that even worth it?
I don't even think legally they can do that.
They can't back out China sales
And then they are conducting China sales
In the quarter
That would be my assumption
That doesn't make any sense to me
So you think they have sold zero China chips
I mean if their guidance
Is saying that they're not assuming
Then it would be
It probably wouldn't be above board
In my opinion
I don't know legally but in my opinion It wouldn't be above board that you would guide that you don't have any China sales when you do, like when you currently do.
I'm not saying that's what they're doing.
I want to make sure people understand.
Couldn't they argue and say like, hey, we don't know if there's going to be new export controls that relimit our ability to ship H20s.
And so as a consequence of that, we're not going to factor into the guidance.
I mean, that would be fair, right?
But what he said, the language they use, though,
is quite specific.
The company has not assumed any H20 shipments
to China in the outlet.
I get that.
That's what I'm saying.
That's what I'm saying, okay?
This quarter, they didn't ship them
because the export controls were in place.
But we found out, what was it, four weeks ago? I don't remember when the White House meeting with NVIDIA was when they announced and said, hey, we've made an H20 compliant trip. And then Trump came out on Truth Social and said saying we're not factoring it into our forward guidance, even though they are now allowed to ship them.
So the only justification for that would be to say, yeah, we're now allowed to ship them under the agreement that was made three weeks ago.
But in lieu of the, our inability to predict
what might happen with export controls,
we're not gonna factor it into guidance.
Like, I don't know if that's-
That might happen.
I'm gonna tell you right now,
like you and I have in this conversation
and everybody else,
there's an analyst that will be asking the question.
That'll be probably one of the first questions
that are asked today, right?
Like we will have clarity about it.
Yeah, cause that is uh you know and
it also could lead to maybe he hypes up this new chip that's going to be china compliant you know
he could be hyping that up as well right i don't know for what it's worth many chinese companies
have data centers outside china huawei has more than a dozen large data centers outside China. So
there's so much fudge in these numbers that they could do a lot with it, but they're being very
conservative, which is probably the right thing at this time. And then the teardown, right? So they
wrote these H20s down. I think some of these parts were also then like shipped out to other
products as well,
if I'm not mistaken. Who knows? There's a lot that's going on with it. I mean, in the grand
scheme of things, their core business is still doing a phenomenal job. So I'm not sure if China
sales is going to be a make or break for this company over the long term. It's just, once again,
over the long term.
It's just, once again,
very conservative guide on the China sales.
Maybe they're not able to ship them now.
Maybe they're just ramping those up.
I just found it very interesting
that they didn't probably elaborate a little bit more.
It seemed like more of a definitive statement
like we saw last quarter.
Well, yeah, I mean,
I think the question is on fairness right like if it's fair for them to
to to exclude that from their guidance and i mean i think it's tough to make the argument
in either direction because it is logical and fair for them to say hey this we have an
unpredictable policy environment surrounding the product.
That's a fair statement for NVIDIA to make.
If I was a company and I had a core product that was being sold in massive quantity to
one specific country and the executive branch of the government can just turn the light
switch on and off at will on my ability to export that product
you know you some people will call it unfair sandbagging but i think it's okay i think it's
okay to say hey we don't know if we'll be able to ship or what quantity we'll be able to ship
of that product to china so i i think nvidia is fine to exclude it to be to be completely honest
but we'll have to hear the details on the call i would agree with you and then also too i mean
they and they pre-announced the write down.
If they do have more definitive terms that they trust,
they'll probably also then pre-announce
any type of adjustments to their guy, right?
I mean, it seems like they're prudent enough.
Yeah, exactly.
Yeah.
Monasif, you were talking there?
No, I'm still trying to figure out
how to break down that uh the the margin this quarter
so so clearly they sold some h20s but that was a release from the you know the write down so leave
that out of the picture it still looks like they're selling a significant amount of uh of
hopper series which is obviously lower margin than Blackwell.
So this is not, we are still not at a Blackwell-centric quarter yet.
So it's very difficult to get your margins above that.
How much are margins up monitored from last quarter?
What's that?
Same, 71 something, 71 point something percent.
Which were they last quarter, do you know?
Or Evan, do you know?
Evan's usually good at finding that information quickly.
Can you hear me?
Yeah, it's 70%.
I don't know.
Yeah, it was 70%.
Yeah, I'm saying, oh, 70% last quarter, and then what was it this quarter?
71%.
It's 71.3 last quarter, 72.3 this quarter.
Okay, so it bumped up a little bit.
But they're not back at highs, right?
What were the highs that they said they were guiding to get back to in the back half?
76, something like 77% was the high.
But you're not going to get to that with the three-year-old product.
And so they're reiterating that, right?
There's no way to have a hopper in the mix and get to that.
It's not possible.
But the language around that guidance changed right monotiv because last quarter with the margin guidance they said we will return to all-time
highs and now they're saying we will return to mid-70s i well they said we'll return to mid-70s
last quarter also but no last quarter they said all-time highs i remember that explicitly because
we talked about that on the spaces they are yeah i mean there's no way it clearly means that they're having some trouble getting as many offloading existing
either they're either they're having trouble ramping blackball at scale slightly incrementally
or they're having trouble offloading previous gen inventory one of the two or both or a little bit
of both but what i'm saying is that that is an incremental change in the language around margin guidance, right? To go from saying
we will be at all time high margins in the back half of the year to now saying we will be at mid
70s. Right? That's a different that's a change in language. And it's not a big change in language,
but it is a change in language, right? Because quarters in a row they said we will nvidia will return to all-time highs in
margins and now they're not saying that you know so i don't know that that that's that's
somewhat notable maybe not a big deal but somewhat notable
it's still unique with their product cycle too, right? Like this is kind of an abnormal product cycle.
So, you know, who knows, is it a demand issue?
Probably not.
Do they have too much supply of the hoppers?
That might be the case.
Maybe there's buyers that want the hoppers still,
like who knows?
Once again, we're gonna get clarity
in whenever this conference call is gonna be,
but it's a very aggressive product cycle
and I keep bringing it up.
It's actually very impressive that they and i keep bringing it up it's
actually very impressive that they're going to continue on with it and the fear of cannibalization
in my opinion is still out there long term i mean the hopper for a lot of use cases for a lot of
these startups that are buying chips the hopper is better than even the competitors, you know, first gen chips. So like, yeah,
they're probably having still a tremendous amount of demand for Hopper and
not as much spill over to Blackwell as they expected or as quickly as they
expected. So that might be part of the issue, but I, it's,
it's hard to say exactly what it is. It could be a number of things, but yeah,
I mean, it's still a good quarter overall. I'm not knocking them.
I'm just saying it is interesting. I think like I don't think they're going to get back
to those margins sustainably now. Maybe they get back there for a quarter, but like.
I don't think so. I said that last quarter, too, and now what we're up a percent on margin. So,
you know, maybe we are headed incrementally higher, but it's to me, it's really hard to
wrap my mind around them sustaining that kind of margin profile, but who knows.
We are getting to the point where we are, what, 20 minutes from the earnings call coming up.
Kevin kind of gave the point a little bit about one of the questions that he thinks is going to be asked right away.
I mean, there's so much uncertainty around what's going
on with these China chip sales.
What even is, what's,
what can we even do right now? So
I think that's definitely one of the questions
that's going to be asked, but I wonder what you guys are expecting
from this earnings call. What type of questions
you want to hear be asked.
Maybe I'll keep it open, see where the discussion goes, but
yeah.
Do they say where their chips are going
if they're not going to China?
Like the chips that are designated for China?
No, they didn't say it.
That's what I was trying to interject there.
My best guess is that you know Chinese companies data centers
outside China are an easy mark for that so I mean if they could just sell to anybody
they would not have taken the charge last quarter right so it probably was a customer
specific problem that they could not find somebody to
offload that to the other possibility is that they found a price where they could offload it
right so they they probably took a you know zero margin on it and to offload it possibly
i'm just guessing here they didn't give that explanation
they did say that they were able to shift some of the sales as you
were saying there so there was some it was the answer wasn't fully none but and
there is that they do give away a revenue by segment I haven't really
heard and even talk about that if I can I can find it I'm still I'm still waiting
for their detailed accounts they've not posted that yet
dav do we have you here
all right no worries i am i i see you i need to sorry i had to change my mic input what's going on you are good oh you're on the computer yeah how's it sound no i was just
one to bring into it it sounds pretty decent
yeah i don't know i mean i've been looking at this the reaction seems a little bit muted i
suppose based on the implied volatility
and move that was talked about beforehand. People are not doing great right now, but
you know, not a little bit less than 3%. I'm mostly curious to see what they say on the
call. How much do you think that for this one, it was what they say on the call versus
what they reported the numbers?
Yeah, I think that, I think they said a lot in those, in the numbers and what some of the questions
are going to be.
But for me, the whole conversation we were having there is the way NVIDIA guides it is
conservatively and leaves you with upside for certain stuff.
So that would be my lean towards it, towards it is what they're going to say is we don't
know.
I think we're just at. The unadjusted eps was a dollar eight so you're
looking at next 12 months of roughly you know four and a half dollars with a little bit of growth
four and a half dollars would make it you know 40 times now we are you know roughly you know, 10%, you know, larger multiple than Microsoft. That's it. It's 40 times forward
is not is is is ridiculous for the kind of growth they still have.
I mean, pretty awesome to see the double beat, though, at the end of the day, and that they continue to be able to...
Gotcha.
Yeah?
All right.
Anyone see the detail numbers?
Can you chime up here, please?
I actually wanted to... Sorry, yeah.
What was the question there?
And then I had a question for Stock Talk.
Can you repeat that?
No, you're good.
You can go.
You can go.
Oh, okay.
He's waiting for the more results.
They normally have, like, four or five things here. i don't is it in the cfo commentary no all right so yeah we're waiting
for a little bit more data on on their balance sheet all that stuff cool but uh yeah yeah question
for stock talk then that i want to pull into this i feel like right now the only bear case around
nvidia and it's hard to have a bear case because they're the clear obvious leader in ai and
networking software services all that stuff is just the geopolitical
tensions, right? And kind of clouding some of the AI business in China. I'm curious when you look at
it, if you see any true competition that is maybe like, I think it was CanberCon or stuff like that,
that are really like posting big numbers and actually have a chance of competing in that region. And also really
interesting spike back up and down there on a video right now, we're down 3.2%. But that was
kind of my question for you, Stock Talk, is if you've looked into that region to see who the
true competition is. Into the China region? Yeah, exactly. No, no, not really. I mean,
I typically don't buy China stocks i mean i have occasionally traded like
alibaba i've occasionally traded like pin duo duo but i i don't like buy and hold china stocks so
no i i haven't bothered do you think there could be in do you think there could be like an adr that
comes out of there that is actually interesting though in this area maybe but i'd rather invest
in u.s data center than i would in china data center to be
honest so no not really there's not there's not one that i would probably i would invest in i mean
maybe a player emerges that ends up getting u.s listed through an adr i don't know if this is the
right climate for that but maybe that happens but uh no there's not any particular players i'm
watching i mean china has their own data center infrastructure players but nobody that happens but uh no there's not any particular players i'm watching i mean china has
their own data center infrastructure players but nobody that i've considered investing in
personally so i wouldn't want to just bs you and talk about because i haven't i haven't researched
it yeah no worries we did see some uh not so great um china macro news yesterday but baba did get a little bit of a pump up
yesterday baba is one of the largest hyperscalers in the entire world uh especially one of the
largest in china they got a little bit of a bid after their largest um silicone manufacturer did
say that they're upping the orders with baba so but yeah i mean it is not nearly as big as the
market cap of a lot of companies
in the US.
So I would attribute China to be a lot of trading per se versus much more heavily focused
on fundamentals.
I mean, there's like a lot of mispriced assets in China.
Like I think Bob is pretty mispriced considering that they are one of the largest hyperskills
in the entire world.
But at the same time, investing for long term is just, it's so difficult to be ex-U.S. when it comes to a lot of trades.
I know there's a lot of other plays when it comes to the non-data center segment outside of the U.S.
But yeah, China just, you don't even know if the data we're hearing from them is actually true.
That's the only problem when it comes to that.
There's also ADR risk with China. So in Trump 1.0, there's a bunch of companies that got
delisted that were trading on US exchanges because there was a claim that they were a part of
Chinese military or the Chinese government. So I think there's also probably a little bit of a
discount for those names that you kind of talked about just because of that perceived risk.
Because at any time, they can be delisted.
I had to work on a project to mitigate some of that stuff, and it's a mess.
So that would be another thing to consider when you're looking at China stocks.
Yeah, but they've made the threat in the past.
And I feel like, I mean, do you think that they're actually going to come in and do that?
Yeah, dude. Trump did it the first time. You like delisted like 30 companies.
Like some of them were actually like big telecoms for China where people had
options and then the OCC had no clue how to settle them out.
And people literally lost premiums. Like if,
even if they were in the money just wiped away. So yeah, there is a risk there.
Well, I mean, that's the whole point of the onshore manufacturing.
We want to de-risk China, right?
We're literally sitting here at risk every single day that China could
invade Taiwan.
I mean, I don't think they're going to, but that's an overhead risk.
I mean, obviously there's a lot more dependent on that.
Like I'm not a macro expert, but like,
if you take away that risk of bringing,
having a lot of manufacturing in Taiwan,
also in China as well,
it'll really take away the position that China might have being the second
most powerful country in the entire world,
according to GDP.
Yeah.
I'm looking forward to this earnings call.
So there's definitely going to be a China question coming out.
And maybe all the questions are going to be around China.
What do you think other more of these analyst questions are going to be like?
Maybe Wolfie, I think we still have you up here.
Maybe I'll throw that question over to you first.
The China question, that's a given.
What do you think is another question we're going to hear an analyst ask Jensen?
I mean, I just think the data center thing is going to be the core focus.
So a bunch of questions around there.
Probably the buyback.
It's a $60 billion buyback.
Probably get some questions around that.
Is it accelerated?
Is it not?
That's kind of like the thing there for me but i don't i don't think there's too much outside of what you guys talked about i think if they've got any any sort of clarity on the blackwell stuff uh when they
expect it to be a bigger segment i think that's kind of cool to know but um no i mean i think obviously it's just you know
it's a four trillion dollar company so i'm not discounting the size of of the dollar amount but
you know it's a three percent move give or take you know there's plenty of days where this thing
moves three percent either direction implied was five percent like this is just this is just loose
change right now still the earnings
call still time you know the one thing that the one thing that i'll say outside of questions the
one thing this is the second quarter where you know the release is kind of muddy you know i don't
remember previous quarters prior to the last two being this muddy where it's like you're trying to
figure out you know where one thing ends where the other thing begins so i think that's from like for me just the antenna goes up a little bit on that
because it used to be pretty straightforward pretty simple just here's your growth here's
this here's that it's on fire it's beating expectations but now it's like you know and i
think that's a that's a i think that's a feature not a bug i think they're trying to navigate this
this uh really strange landscape where
things can change on a dime
and there's a lot of quid pro quo
going on. So I think they're doing
I think
they're doing it as best they can.
But I do
find it kind of interesting. This is the
second quarter where it's kind of muddy and we're
trying to piece together
where the H20 thing begins and ends, how much is to china where is it going to etc etc so i'd pay
attention to that moving forward too
yeah i pulled up the sec filing so there's a little more data. So they reversed 180 million out of that 5.4 billion charge.
So other interesting things, their day sales outstanding jumped from 46 to 54. Now that's not
good at the, you know, just off the top. It's not a good thing. But they're saying that it's driven
by timing of cash collection. Everybody says that so I would take that with
a pinch of salt. But they're also seeing that Blackwell Ultra was ramping only late in the
quarter. So that was timing of that. Inventory is higher again for to support the Blackwell
ramp, Blackwell Ultra ramp. So inventory went up almost 4 billion. So that's a lot. Purchase commitments also up 2.5 billion. Again on,
you know, supporting, you know, Blackwell Ultra launch and, you know, and additional
manufacturing capacity they needed for that. So they're still grabbing more capacity. So
it's not, it's not like there's anything here that explains a slowdown, but
there are some weak spots that we have to be concerned with here. So anyway, I'm still reading.
I'll come back as I read more. So data center re-accelerated from last quarter. I'm seeing here
that, I'm sorry, not data center, gaming.
This was actually something that I was kind of speculating to happen. They grew 42%.
0.2% year over year for gaming last quarter. This quarter, they grew 48.9%. And this is about 10% of revenue. So this could have a meaningful impact that this continues to re-accelerate
into next quarter. but they did guide
for our data center came in at which is 90 percent about 90 percent revenue came in at 56.4 percent
so we could make up for a little bit of slack there if that continues uh automotive was 72.3
last quarter came in in 69 so it's about about right there. Professional visualization.
I mean, these other segments are negligible,
but gaming segment, if this continues to accelerate,
which we're looking at AMD as well,
looking at other gaming companies,
this could pick up the slack when it comes to NVIDIA.
I don't think this should be ignored.
So again, staying there, Sam,
within data center, right,
compute was down sequentially quarter over quarter.
It was down a couple of hundred million.
Now that could very well be just the uncertainty or any China sales of data center products, right?
So that could be it.
But networking is up 50%, 46% quarter over quarter
and 100% year over year, right?
So that is a read-through for other companies. That means you know, they
they're, you know, grabbing market share somewhere. It's
no seven and a half billion. So it's not a small number anymore.
So these guys are grabbing network market share. So, so
it's a it's a it's a total solution sale, which they're
going on with Blackwell, right? And you're gonna see more as
Blackwell, relative to how much of opera is sold when they see larger blackwell sales you're going
to see that network revenue also start to you know become more and more significant i mean
out of the 40 41 billion 7.2 billion is already networking networking listen
I'm sorry we still might be
green a lot of Jensen
Wong left for the record we are
going to listen to this earnest call here for the next hour
after this Jensen
generally goes on a lot of mainstream media
and will do multiple interviews.
So keep an eye out for that as well.
I don't think we'll stay live for all of that, but Bloomberg, CNBC, I know Ali was on.
I will say, in spite of us kind of looking at the potential weaknesses and risks,
it is impressive that they're posting these kind of margins still and these kind of numbers still,
considering the competitive environment just in the last six months has gotten a lot better.
We have to know at some point NVIDIA will have competition.
At some point.
And in the last six months, that possibility has become incrementally higher.
And probably in the next six months, it has become incrementally higher and probably in the next six
months it'll become incrementally higher again and eventually that means margin risk which is
probably the first thing that you'll see an inability to return to those all-time high
margins which i don't think is going to happen regardless of competition and second secondly
you'll see you know this crazy expansion we've seen in data center revenue and other parts potentially slow down a little bit.
Now, today's data center miss is not a big miss.
It's a couple hundred million bucks, but it is the first data center miss since Q1 of 2023.
So is that a sign or whatever?
I don't know.
But that is worth noting, I guess.
And eventually it can compound into a bigger risk if that slowdown becomes more apparent.
And if on top of that, there is also an inability to return to all-time high margins, which I think is going to happen either way.
Or it's not going to happen either way.
So what do we think about nebius man
but i still obviously don't like nebius but nebius is probably get going to get hit tomorrow if
nvidia stays red um as are most of the data center names are probably going to see some
selling action as a consequence of that but i think nebius you know uniquely has other exposures
outside of data center that make a lot a lot value, like ClickHouse, which is my favorite,
one of their subsidiaries, but AVRide as well.
The thing, though, is on NVIDIA,
with the data center business,
we know that it's supply constrained.
So I don't know what you necessarily take away from this.
For now, but this miss can't be anything else besides that.
We know there's supply constrained.
Yeah, for now there's supply constrained. Yeah, for now, there's supply constraint.
Yeah, I agree.
I'm not, like, again, this was a great quarter.
All things being said, this was a great quarter.
It's a great business.
But that will happen eventually.
Like, eventually, I mean, and this,
it doesn't just have to happen by virtue of competition.
It doesn't have to happen by virtue of volume growth slowing.
No, yeah, obviously. It could happen by virtue of those two things. But to happen by virtue of volume growth slowing no yeah it
could happen it could happen by virtue of those two things but i think there are other ways the
law of big numbers history would have told you this would have happened already a couple quarters
ago there's never been a move like this that doesn't mean that's not going to happen just
because it hasn't it's going to happen it will happen it's going to happen yeah what i what but
it's not it doesn't just have to be by virtue of competition or volume growth selling. It can also be by virtue
of strategy changing in the industry, which may happen as well. Like eventually we may realize
that like once the initial stage one and stage two of the data center build out are complete,
that the training component of that build out may not be as necessary as the inference component of
that build out. And that may affect the growth of deployment, right? Because if right now, build out is serving two
purposes, it's serving the purpose of training new models, and it's also servicing the purpose
of inference for existing models. At some juncture, that balance will change, right? Just
think about it logically. Like at at some point we will have very sophisticated
models that have already been trained and we will yes still need plenty of compute for inference
but it will change the pace of build out right it's just like logic 101 isn't a data center
expert to make that yeah yeah so when that happens n NVIDIA's margins will inevitably see compression. I just don't know when that's going to be.
NVIDIA stock would keep rip roaring for the next 10 years. I just don't know when that inflection point is going to happen, but it will happen.
Like no industry that happens in every industry eventually. how long you can keep that buffer of innovation that nvidia has where they're three generations
ahead of the competition and basically just dominate the industry all the biggest customers
buy from them eventually that'll change and it may not change in the sense that they'll go to
a competitor but it may change in the sense that they just won't need as okay there is a headline
right now u.s government has not pet so it's the i think it's penned regulation codifying the 15
so i think that that whole 15 yeah u.s government officials express
expectations that the u.s government will receive 15 of revenue from license
express expectations so it sounds like that is not has not got into effect is what this is all saying
let me pull this up.
But I think that, and the same from an SEC file.
NVIDIA put it in their 10 queue.
Yeah, let me pull this up.
Yeah, okay, that's fine. That's okay.
Look, I have no disagreement with anything Stock Talk said and I'll go one step further.
It's not just margin compression, right? Revenues could go down dramatically too. So that risk is always there and it's not that far away because we know that revenue from AI has not, you know, kept pace with the normal
return on invested capital of tech. So there is a cost to pay here. And at some point in time,
you know, you can't keep going at this pace. So that'll change. But at least for immediate quarter,
But at least for immediate quarter, what I am enthusiastic about is the fact that they are reserving ever larger capacity and holding ever larger inventory of parts to service a growing revenue.
So for now, unless they are completely wrong about this, next quarter will be better than this quarter at least marginally if not
if not if not by a huge number and we're talking very large numbers here already right so
so the the fact that they've reserved more capacity this quarter than uh i mean next
quarter than come this current quarter that that's a positive thing
yeah i mean look the company is it's a fantastic company. Yeah, I mean, look, the company is,
it's a fantastic company.
They're still executing.
But like I said, I just think, you know,
every market leader, 75% margin story
eventually gets either partially commoditized
or something happens.
I just went over like 12 factors earlier,
but one of those 12 things that I mentioned could happen.
And, you know, it changes the entire industry.
It's not just going to be a thing for industry.
The call started, so I'll go listen and I'll come back with some updates.
Yeah, Evan, I don't know if you wanted to play the call.
No, I was.
They were doing the operator giving their stuff.
So we have a second before we get anything, but the earnings call is started.
If you would like to withdraw your question.
Now listen, it's through an audio.
Sounds fine.
Clear?
So Shihari, you may begin your conference.
Good afternoon, everyone.
And welcome to NVIDIA's conference call for the second quarter of fiscal 2026.
With me today from NVIDIA are Jensen Wong, President and Chief Executive Officer, and
Colette Kress, Executive Vice President and Chief Financial Officer.
I'd like to remind you that our call is being webcast live
on NVIDIA's Investor Relations website.
The webcast will be available for replay
until the conference call to discuss our financial results
for the third quarter of fiscal 2026.
The content of today's call is NVIDIA's property.
It can't be reproduced or transcribed
without our prior written consent.
During this call, we may make forward-looking
statements based on current expectations. These are subject to a number of significant risks and
uncertainties and our actual results may differ materially. For a discussion of factors that could
affect our future financial results in business please refer to disclosure in today's earnings
release, our most recent forms 10K and 10Q, and the reports
that we may file on Form 8K with the Securities and Exchange Commission. All our statements are
made as of today, August 27th, 2025, based on information currently available to us.
Except as required by law, we assume no obligation to update any such statements.
During this call, we will discuss non-GAAP financial measures.
You can find a reconciliation
of these non-GAAP financial measures
to GAAP financial measures in our CFO commentary,
which is posted on our website.
With that, let me turn the call over to Collette.
Thank you, Toshia.
We delivered another record quarter
while navigating what continues to be a dynamic external environment.
Total revenue was $46.7 billion, exceeded our outlook as we grew sequentially across all market platforms.
Data center revenue grew 56% year over year.
Data center revenue also grew sequentially despite the $4 billion decline
in H20 revenue. NVIDIA's Blackwell platform reached record levels, growing sequentially by
17%. We began production shipments of GB300 in Q2, our full- full stack AI solutions for cloud service providers, neoclouds, enterprises,
and sovereigns are all contributing to our growth. We are at the beginning of an industrial
revolution that will transform every industry. We see three to four trillion dollars in AI
infrastructure spend by the end of the decade.
The scale and scope of these build-outs
present significant long-term growth opportunities
for NVIDIA.
The GB200 NBL system is seeing widespread adoption
with deployments at CSPs and consumer internet companies.
Lighthouse model builders, including OpenAI, Meta,
and TheStraw are using the GB200 NBL72
at data center scale for both training
next generation models
and serving inference models in production.
The new Blackwell Ultra platform
has also had a strong quarter,
generating tens of billions in revenue.
The transition to the GB300 has been seamless
for major cloud service providers
due to its shared architecture, software,
and physical footprint with the GB200,
enabling them to build and deploy GB300 racks with ease.
The transition to the new GB300 rack-based architecture
has been seamless.
Factory builds in late July and early August
were successfully converted to support the GB300 ramp.
And today, full production is underway.
The current run rate is back at full speed,
producing approximately
1,000 racks per week. This output is expected to accelerate even further throughout the third
quarter as additional capacity comes online. We expect widespread market availability in the
second half of the year as CoreWeave prepares to bring their GV300 instance to market as they are
already seeing 10x more inference performance on reasoning models compared to H100. Compared to
the previous hopper generation, GV300 and VL72 AI factories promise a 10x improvement in token per
watt energy efficiency,
which translates to revenues as data centers are power limited.
The chips of the Rubin platform are in fab.
The Vera CPU, Rubin GPU, CX9 SuperNIC,
NVLink 144 scale-up switch,
Spectrum X scale-out and scale-across switch,
and the silicon photonics processor.
Rubin remains on schedule for volume production next year.
Rubin will be our third-generation NVLink
rack-scale AI supercomputer
with a mature and full-scale supply chain.
This keeps us on track with our pace
of an annual product cadence
and continuous innovation across compute,
networking, systems, and software.
In late July, the U.S. government
began reviewing licenses for sales of H20
to China customers.
While a select number of our China-based customers have received licenses over the past few weeks,
we have not shipped any H-20 based on those licenses.
an expectation that the USG will receive 15% of the revenue generated from licensed H20 sales.
But to date, the USG has not published a regulation codifying such requirement.
We have not included H20 in our Q3 outlook as we continue to work through geopolitical issues. If geopolitical issues reside,
we should ship $2 to $5 billion in H20 revenue in Q3. And if we add more orders, we can bill more.
We continue to advocate for the U.S. government to approve Blackwell for China. Our products are designed and sold
for beneficial commercial use,
and every licensed sale we make
will benefit the U.S. economy,
the U.S. leadership.
In highly competitive markets,
we want to win the support of every developer.
America's AI technology stack can be the world's standard
if we race and compete globally.
Notably in the quarter was an increase in Hopper 100 and H200 shipments.
We also sold approximately $650 million of H20 in Q2 to an unrestricted customer outside of China.
The sequential increase in Hopper demand indicates
the breadth of data center workloads that run on accelerated computing and the power of CUDA
libraries and full stack optimizations, which continuously enhance the performance and economic
value of our platform. As we continue to deliver both Hopper and Blackwell GPUs, we are focusing on meeting the soaring global demand.
This growth is fueled by capital expenditures from the cloud to enterprises,
which are on track to invest $600 billion in data center infrastructure
and compute this calendar year alone, nearly doubling in two years.
We expect annual AI infrastructure investments to continue growing this calendar year alone, nearly doubling in two years.
We expect annual AI infrastructure investments to continue growing driven by the several factors,
reasoning agentic AI requiring orders of magnitude
more training and inference compute,
global build outs for sovereign AI,
enterprise AI adoption,
and the arrival of physical AI and robotics.
Blackwell has set the benchmark as it is the new standard for AI inference performance.
The market for AI inference is expanding rapidly with reasoning and agentic AI gaining traction across industries.
Blackwell's rack scale NVLink and CUDA full stack architecture addresses this by redefining the economics of inference.
New NV, FP4, 4-bit precision and NVLink 72
on the GB300 platform delivers a 50x increase in energy efficiency per token
compared to Hopper, enabling companies to monetize their compute at unprecedented scale.
For instance, a 3 million investment in GB200 infrastructure can generate
30 million in token revenue, a 10x return.
NVIDIA's software innovation,
combined with the strength of our developer ecosystem,
has already improved Blackwell's performance
by more than 2x since its launch.
Advances in CUDA, Tensor TLLM, and Dynamo
are unlocking maximum efficiency.
CUDA library contributions from the open source community
along with NVIDIA's open libraries and frameworks
are now integrated into millions of workflows.
This powerful flywheel of collaborative innovation
between NVIDIA and global community contribution
strengthens NVIDIA's performance leadership.
NVIDIA is a top contributor to open AI models,
data, and software.
Blackwell has introduced a groundbreaking numerical approach to large language model free training.
Using NVFP4, computations on the GB300 can now achieve 7x faster training than the H100, which uses FP8.
This innovation delivers the accuracy of 16-bit precision with the speed and efficiency of 4-bit, setting a new standard for AI factor efficiency and scalability.
The AI industry is quickly adopting this revolutionary technology with major players
such as AWS, Google Cloud, Microsoft Azure, and OpenAI, as well as Coquire, Mastral,
Kimi AI, Perplexity, Reflection, and Runway, already embracing it.
NVIDIA's performance leadership was further validated in the latest MLPerth training benchmarks,
where the GB200 delivered a clean sweep.
Be on the lookout for the upcoming MLPerf inference results in September, which will include
benchmarks based on the Blackwell Ultra. NVIDIA RTX Pro servers are in full production for the
world's system makers. These are air-cooled PCIe-based systems integrated seamlessly into standard IT environments and run traditional
enterprise IT applications, as well as the most advanced agentic and physical AI applications.
Nearly 90 companies, including many global leaders, are already adopting RTX Pro servers.
Hitachi uses them for real-time simulation and digital twins, Lily for drug
discovery, Hyundai for factory design and AV validation, and Disney for immersive storytelling.
As enterprises modernize data centers, RTX pro servers are poised to become a multi-billion
dollar product line. Sovereign AI is one on the rise as the nation's ability to
develop its own AI using domestic infrastructure data and talent presents a significant opportunity
for NVIDIA. NVIDIA is at the forefront of landmark initiatives across the UK and Europe. The European Union plans to invest 20 billion euros
to establish 20 AI factories across France, Germany, Italy, and Spain, including five
gigafactories, to increase its AI compute infrastructure by tenfold. In the UK, the Umbar'd AI supercomputer powered by NVIDIA was unveiled at the country's most powerful AI system, delivering 21 exaplots of AI performance to accelerate breakthroughs in fields of drug discovery and climate modeling.
We are on track to achieve over $20 billion in sovereign AI revenue this year,
more than double than that of last year. Networking delivered record revenue of $7.3 billion
and escalating demands of AI compute clusters necessitate high efficiency and low latency networking. This represents a 46% sequential
and 98% year on year increase
with strong demand across Spectrum X Ethernet,
InfiniBand and NVLink.
Our Spectrum X enhanced Ethernet solutions
provide the highest throughput and lowest latency network for Ethernet AI workloads.
Spectrum X Ethernet delivered double-digit sequential and year-over-year growth
with annualized revenue exceeding $10 billion. At Hot Chips, we introduced Spectrum XGS Ethernet,
a technology designed to unify disparate data centers into gigascale AI super
factories. For Weave is an initial adopter of the solution, which is projected to double
GPU to GPU communication speed. InfiniBand revenue nearly doubled sequentially, fueled by the adoption of XDR technology, which provides double the bandwidth improvement over its predecessor, especially valuable for the model builders.
The world's fastest switch, NVLink, with 14x the bandwidth of PCIe Gen 5 delivered strong growth as customers deployed race Blackpool
and V-Link rack scale systems. The positive reception to NV-Link Fusion, which allows
semi-custom AI infrastructure, has been widespread. Japan's upcoming Fugaku Next will integrate Fujitsu's CPUs with our architecture via NVLink Fusion.
It will run a range of workloads, including AI, supercomputing, and quantum computing.
Fugaku Next joins a rapidly expanding list of leading quantum supercomputing and research centers running on NVIDIA's Q2Q quantum platform,
including ULIC, AIST, NNF, and NERS, supported by over 300 ecosystem partners, including AWS,
Google Quantum AI, Quantinium, QERA, and PsiQuantum.
Just in Thor, our new robotics computing platform is now available.
Thor delivers an order of magnitude greater AI performance and energy efficiency than NVIDIA AGXORN.
It runs the latest generative and reasoning AI models at the edge in real time, enabling state-of-the-art robotics.
Adoption of NVIDIA's robotics full-stack platform is growing at rapid rate.
Over 2 million developers and 1,000-plus hardware, software applications and sensor partners taking our platform to market.
applications and sensor partners taking our platform to market. Leading enterprises across
industries have adopted Thor, including Agility Robotics, Amazon Robotics, Boston Dynamics,
Caterpillar, Figure, Hexagon, Medtronic, and Meta. Robotic applications require exponentially
more compute on the device and in infrastructure representing
a significant long-term demand driver for our data center platform.
NVIDIA Omniverse with Cosmos is our data center physical AI digital term platform built for
development of robot and robotic systems.
This quarter, we announced a major expansion of our partnership with Siemens to enable AI, automatic factories, leading European robotics companies, including Agile Robots, Neuro Robotics, and Universal Robots, are building their latest innovations with the Omniverse platform.
with the Omniverse platform.
Transitioning to a quick summary of our revenue by geography.
China declined on a sequential basis
to low single digits percentage of data center revenue.
Note, our Q3 outlook does not include H20 shipments
to China customers.
Singapore revenue represented 22%
of second quarter's billed revenue
as customers have centralized
their invoicing in Singapore. Over 99% of data center compute revenue billed to Singapore
was for US-based customers. Our gaming revenue was a record, $4.3 billion, a 14% sequential
increase and a 49% jump year-on-year. This was driven by the ramp of
Blackwell GeForce GPUs as strong sales continued as we increased supply availability. This quarter
we shipped GeForce RTX 5060 desktop GPU. It brings double the performance along with advanced
ray tracing, neural rendering, and AI-powered DLSS4 gameplay to millions of gamers worldwide.
Blackwell is coming to GeForce Now in September.
This is GeForce Now's most significant upgrade, offering RTX 5080 class performance, minimal latency, and 5k resolution at 120 frames per second. We are also doubling
the GeForce Now catalog to over 4,500 titles, the largest library of any cloud gaming service.
For AI enthusiasts, on-device AI performs the best RTX GPUs. We partnered with OpenAI to optimize their open source GPT models for
high quality, fast and efficient inference on millions of RTX enabled window
devices. With the RTX platform stack,
window developers can create AI applications designed to run on the world's
largest AI PC user base.
Professional visualization revenue reached $601 million, a 32% year-on-year increase.
Growth was driven by an adoption of the high-end RTX workstation GPUs and AI-powered workload
like design, simulation, and prototyping.
Key customers are leveraging our solutions
to accelerate their operations.
Activision Blizzard uses RTX workstations
to enhance creative workflows,
while robotics innovator figure AI
powers its humanoid robots with RTX embedded GPUs.
Automotive revenue, which includes
only in-car compute revenue, was $586 million, up 69% year-on-year, primarily driven by self-driving solutions.
We have begun shipments of NVIDIA Thor SoC, the successor to Orin.
Thor's arrival coincides with the industry's accelerating shift to vision, language, model architecture, generative AI, and higher levels of autonomy.
Thor is the most successful robotics and AV computer we've ever created, Thor Willpower.
Our full stack drive AV software platform is now in production,
platform is now in production, opening up billions to new revenue opportunities for
NVIDIA while improving vehicle safety and autonomy. Now moving to the rest of our P&L.
GapRose margin was 72.4% and non-GapRose margin was 72.7%. These figures include a 180 million or 40 basis point benefit from releasing previously reserved H20 inventory.
Excluding this benefit, non-GAAP gross margins would have been 72.3%, still exceeding our outlook.
GAAP operating expenses rose 8% and 6% on a non-GAAP basis sequentially.
This increase was driven by higher compute and infrastructure costs,
as well as higher compensation and benefit costs.
To support the ramp of Blackwell and Blackwell Ultra,
inventory increased sequentially
from 11 billion to 15 billion in Q2.
While we prioritize funding, our growth and strategic billion in Q2. While we prioritize funding our growth and strategic initiatives
in Q2, we returned 10 billion to shareholders through share repurchases and cash dividends.
Our board of directors recently approved a 60 billion share repurchase authorization
to our remaining 14.7 billion of authorization at the end of Q2. Okay, let me turn
it to the outlook for the third quarter. Total revenue is expected to be 54 billion plus or minus
2%. This represents over 7 billion in sequential growth. Again, we do not assume any H-20 shipments to China customers in our outlook. Gap and non-gap
gross margins are expected to be 73.3%, 73.5% respectively, plus or minus 50 basic points.
We continue to expect to exit the year with non-gap gross margins in the mid 70s. GAP and non-GAAP operating expenses are expected
to be approximately 5.9 billion and 4.2 billion respectively. For the full year we expect operating
expenses to grow in the high 30s range year over year up from our prior expectations of the mid 30s.
We are accelerating investments in the business
to address the magnitude of growth opportunities that lie ahead. Gap and non-gap other income and
expenses are expected to be an income of approximately $500 million, excluding gains
and losses from non-marketable and public held equity securities. Gap and non-gap tax rates are expected to be 16.5%, plus or minus 1%,
excluding any discrete items. Further financial data are included in the CFO commentary and other
information available on our website. In closing, let me highlight upcoming events for the financial
community. We will be at the Goldman Sachs Technology Conference on September 8th in San Francisco.
Our annual NDR will commence the first part of October.
GTC data center begins on October 27th
with Jensen's keynote scheduled for the 28th.
We look forward to seeing you at these events.
Our earnings call to discuss the results
of our third quarter of fiscal 2026
is scheduled for November 19th.
We will now open the call for questions.
Operator, would you please pull for questions?
At this time, I would like to remind everyone
in order to ask a question, press star, send the number one on your telephone keypad.
We'll pause for just a moment to compile the Q&A roster.
As a reminder, please limit yourself to one question.
Your first question comes from CJ Muse with Cantor Fitzgerald.
Your line is open.
Yes, good afternoon.
Thank you for taking the question.
I guess with Wafer in to rack out lead times of 12 months,
you confirmed on the call today that Ruben is on track for ramping the second half.
And obviously, many of these investments are multi-year projects contingent upon power,
cooling, et cetera. I was hoping perhaps you take a high-level view and speak to your vision for growth into 2026.
And as part of that, if you can kind of comment between networking data center, it would be very helpful.
Thanks, CJ.
At the highest level of growth drivers would be the evolution, the introduction, if you will, of reasoning agentic AI.
You know, where chatbots used to be one shot, you give it a prompt and it would generate the answer.
Now the AI does research, it thinks and does a plan, and it might use tools.
And so it's called long thinking.
And the longer it thinks, oftentimes it produces better answers.
And the amount of computation necessary for one shot versus reasoning agentic AI models
could be a hundred times, a thousand times, and potentially even more
as the amount of research and basically reading and comprehension that it goes off to do.
And so the amount of computation that has resulted in agentic AI has grown tremendously.
And of course, the effectiveness has also grown tremendously. Because of agentic AI, the amount of hallucination has dropped significantly.
You can now use tools and perform tasks.
Enterprises have been opened up.
As a result of agentic AI and vision language models, we now are seeing a breakthrough in physical AI, in robotics, autonomous systems.
So the last year, AI has made tremendous progress
and agentic systems, reasoning systems
is completely revolutionary.
Now we built the Blackwell MVLink 72 system,
a rack scale computing system for this moment.
We've been working on it for several years.
This last year we transitioned from MVLink 8, which is a node scale computing, each node
is a computer, to now MVLink 72 where each rack is a computer. That this aggregation
of MVLink 72 into a Rack scale system was extremely hard to do, but the results are
extraordinary. We're seeing orders of magnitude speed up and therefore energy efficiency and
therefore cost effectiveness of token generation because of MVLink 72. And so over the next couple of years, you're going to, well, you asked about longer term.
Over the next five years, we're going to scale into with Blackwell, with Rubin, and follow-ons to scale into effectively a $3 to $4 trillion AI infrastructure opportunity.
$4 trillion AI infrastructure opportunity.
The last couple of years, you have seen that CapEx has grown in just the top four CSPs
by has doubled and grown to about $600 billion.
So we're in the beginning of this build out and the AI technology advances has really
enabled AI to be able to adopt and solve problems to many different industries.
Your next question comes from Vivek Arya with Bank of America Securities. Your line is open.
Thanks for taking my question. I just wanted to clarify the $2 to $5 billion in China, what needs to happen, and what is
the sustainable pace of that China business as you get into Q4?
And then, Jensen, for you, on the competitive landscape, several of your large customers
already have or are planning many ASIC projects.
I think one of your ASIC competitors brought on signal that they could grow their AI
business almost 55-60% next year. Any scenario in which you see the market moving more towards
ASICs and away from NVIDIA GPU, just what are you hearing from your customers? How are they managing
this split between their use of merchant silicon and ASICs? Thank you.
Multrin, Silicon, and ASIC.
Thanks, Vidak.
So let me first answer your question regarding what will it take for the H20s to be shipped.
There is interest in our H20s.
There is the initial set of licenses that we received.
And then additionally, we do have supply that we are ready.
And that's why we communicated that somewhere
in the range of about two to five billion this quarter,
we could potentially ship.
We're still waiting on several of the geopolitical issues
going back and forth between the governments
and the companies trying to determine their purchases
and what they want to do.
So it's still open at this time, and we're not exactly sure what that full amount will be that this quarter.
However, if more interest arrives, more licenses arrives, again, we can also still build additional H20 and ship more as well.
NVIDIA builds very different things in ASICs.
Well, let's talk about ASICs first.
A lot of projects are started.
Many startup companies are created.
Very few products go into production.
And the reason for that is it's really hard.
Accelerated computing is unlike general purpose computing.
You don't write software and just compile it into a processor.
Accelerated computing is a full stack code design problem.
And AI factories in the last several years has become so much more complex
because of the scale of the problems have grown so significantly.
It is really the ultimate, the most extreme computer science
problem the world's ever seen, obviously.
And so the stack is complicated.
The models are changing incredibly fast
from generative based on autoregressive
to generative based on diffusion to mixed models
to multimmodality.
The number of different models that are coming out that are either derivatives of transformers
or evolutions of transformers is just daunting. One of the advantages that we have is that NVIDIA is available in every cloud. We're available from every computer company. We're available from the cloud to on-prem to edge to robotics on the same programming model.
And so it's sensible that every framework in the world supports NVIDIA.
When you're building a new model architecture, releasing it on NVIDIA's most sensible. And so the diversity of our platform, both in the ability to evolve
into any architecture, the fact that we're everywhere, and also we accelerate the entire
pipeline. You know, everything from data processing to pre-training to post-training with reinforcement
learning, all the way out to inference. And so when you build a data center with NVIDIA platform in it, the utility of it is best. The lifetime usefulness
is much, much longer. And then I would just say that in addition to all of that,
it is just a really extremely complex systems problem anymore.
You know, people talk about the chip itself.
There's one ASIC, the GPU that many people talk about.
But in order to build Blackwell, the platform, and Ruben, the platform, we had to build CPUs that connect fast memory,
and Ruben the platform, we had to build CPUs
that connect fast memory, low,
extremely energy efficient memory
for large KV caching necessary for agentic AI,
to the GPU, to a super NIC,
to a scale up switch we call MVLink,
completely revolutionary when we're in our fifth generation
now, to a scale out switch,
whether it's quantum or Spectrum X Ethernet, to now scale across switches so that we can prepare
for these AI super factories with multiple gigawatts of computing all connected together.
We call that Spectrum XGS. We just announced that at Hot Chips
this week. And so the complications, the complexity of everything that we do
is really quite extraordinary. It's just on a really, really extreme scale now.
And then lastly, if I could just say one more thing, you know, we're in every cloud for a good reason.
Not only are we the most energy efficient, our perf per watt is the best of any computing platform.
And in a world of power limited data centers, perf per watt drives directly to revenues.
And, you know, you've heard me say before that in a lot of ways the more you buy the more you grow and because our per per dollar the performance per dollar is so incredible you also have extremely
great margins so the growth opportunity with NVIDIA's architecture and the gross margins
opportunity with NVIDIA's architecture is absolutely the best.
And so there's a lot of reasons why NVIDIA is chosen by every cloud and every startup
and every computer company.
We're, you know, really a holistic full stack solution for AI factories.
solution for AI Factories.
Your next question comes from Ben Reitzes with Melius.
Your line is open.
Ben Reitzes Hey, thanks a lot.
Jensen, I wanted to ask you about your $3 to $4 trillion in data center infrastructure
spend by the end of the decade.
Previously, you talked about something in the
$1 billion range, which I believe was just for compute by 2028. If you take past comments,
$3 to $4 trillion would imply maybe $2 billion plus in compute spend. And I just wanted to know
if that was right, and that's what you're seeing by the end of the decade. And I'm wondering what
you think your share will be of that. Your share right now of total infrastructure compute-wise is very high. So
wanted to see. And also if there's any bottlenecks you're concerned about, like power,
to get to the $3 trillion to $4 trillion. Thanks a lot.
Yeah, thanks. As you know, the capex of just the top four hyperscalers has doubled in two years.
As the AI revolution went into full steam, as the AI race is now on,
the CapEx spend has doubled to $600 billion per year. There's five years between now and the end of the decade.
And $600 billion only represents the top four hyperscalers.
We still have the rest of the enterprise companies building on-prem.
You have cloud service providers building around the world.
United States represents about 60% of the world's compute.
And over time, you would think that artificial intelligence would reflect GDP scale and growth.
And so, and would be, of course, accelerating GDP growth. And so, our contribution to that
is a large part of the AI infrastructure.
Out of a gigawatt AI factory,
which can go anywhere from 50 to, you know, plus or minus 10%,
50 to plus or minus 10%, let's say 50 to 60 billion,
let's say 50 to 60 billion,
we represent about 35 plus or minus of that.
And 35 out of 50 or so billion dollars
for a gigawatt data center.
And of course, what you get for that is not a GPU.
I think people were famous for building the GPU and inventing the GPU,
but as you know, over the last decade, we've really transitioned to become an AI infrastructure
company. It takes six chips just to build, six different types of chips just to build an AI,
a Rubin AI supercomputer.
And just to scale that out to a gigawatt,
you have hundreds of thousands of GPU compute nodes
and a whole bunch of racks.
And so we're really an AI infrastructure company.
And we're hoping to continue to contribute to growing this industry, making AI more useful,
and then very importantly, driving the performance per watt.
Because the world, as you mentioned, limiters, it will always likely be power limitations
or AI infrastructure or AI building limitations.
And so we need to squeeze as much out of that factory as possible.
NVIDIA's performance per unit of energy used drives the revenue growth of that factory.
It directly translates.
If you have a 100 megawatt factory, per 100 megawatt drives your revenues.
It's tokens per 100 megawatts of factory.
In our case, also, the performance per dollar spent is so high that your gross margins are also the best.
But anyhow, these are the limiters going forward.
And $3 to $4 trillion is fairly sensible for the next five years.
Next question comes from Joe Moore of Morgan Stanley. Your line is open.
Great. Thank you. Congratulations on reopening the China opportunity. Can you talk about
the long-term prospects there? You've talked about half of the AI software world being
there. How much can NVIDIA grow in that business and how important is it that you get the Blackwell
architecture ultimately licensed there?
The China market, I've estimated to be about $50 billion of opportunity for us this year.
If we were able to address it with competitive products, and if it's $50 billion this year,
million dollars this year, you would expect it to grow, say, 50% per year as the rest
of the world's AI market is growing as well.
It is the second largest computing market in the world, and it is also the home of AI
AI researchers, about 50% of the world's AI researchers are in China.
researchers.
About 50% of the world's AI researchers are in China.
The vast majority of the leading open source models are created in China.
And so it's fairly important, I think, for the American technology companies to be able to address that market.
And open source, as you know, is created in one country, but it's used all over the world.
The open source models that have come out of China are really excellent.
DeepSeek, of course, gained global notoriety.
Q1 is excellent.
Kimi is excellent.
There's a whole bunch of new models that are coming out.
coming out. They're multimodal, they're like great language models and it's really fueled
They're multimodal.
They're great language models.
the adoption of AI in enterprises around the world because enterprises want to build their own custom
proprietary software stacks. And so open open source model is really important for enterprise,
it's really important for SaaS who also would like to build
proprietary systems. It has been really incredible for robotics around the world. And so open source
is really important and it's important that the American companies are able to address it.
This is going to be a very large market. We're talking to the administration about the importance of
American companies to be able to address the Chinese market and as you know H20 has been approved
for companies that are not on the entities list and many licenses have been approved and
and many licenses have been approved.
And so I think the opportunity for us to bring Blackwell
to the China market is a real possibility.
And so we just have to keep advocating the sensibility of
and the importance of American tech companies
to be able to lead and win the AI race
and help make the American tech stack the global standard.
Your next question comes from the line of Aaron Rakers with Wells Fargo. Your line is open.
Wells Fargo. Your line is open.
Yeah, thank you for the question.
You know, I want to go back to the Spectrum XGS announcement this week and, you know,
thinking about the Ethernet product, you know, pushing over $10 billion of annualized revenue.
You know, Jess, what is the opportunity set that you see for Spectrum XGS?
Do we think about this as kind of the data center interconnect layer?
Any thoughts on the sizing of this opportunity
within that Ethernet portfolio?
We now offer three networking technologies.
One is for scale up, one is for scale out,
and one for scale across.
Scale up is so that we could build the largest possible virtual GPU, the virtual compute node.
NVLink is revolutionary.
NVLink 72 is what made it possible for Blackwell to deliver such an extraordinary generational jump over Hopper's NVLink 8.
such an extraordinary generational jump over Hopper's NVLink 8. At a time when we have
long thinking models, agentic AI reasoning systems, the NVLink basically amplifies the
memory bandwidth which is really critical for reasoning systems. And so MVLink 72 is fantastic. We then scale out with networking,
which we have two. We have InfiniBand, which is unquestionably the lowest latency,
the lowest jitter, the best scale-out network. It does require more expertise in managing those networks. And for supercomputing, for the leading model makers,
InfiniBand, Quantum InfiniBand is the unambiguous choice.
If you were to benchmark an AI factory,
the ones with InfiniBand are the best performance.
For those who would like to use Ethernet
because their whole data center is built with Ethernet, we have a new type of Ethernet called Spectrum Ethernet.
Spectrum Ethernet is not off the shelf.
It has a whole bunch of new technologies designed for low latency and low jitter and congestion control.
And it has the ability to come closer, much, much closer to InfiniBand than anything that's out there.
And we call that Spectrum X Ethernet.
And then finally, we have Spectrum XGS, a gigascale for connecting multiple data centers,
multiple AI factories into a super factory, a gigantic system.
into a super factory, a gigantic system.
And you're going to see that networking, obviously,
is very important in AI factories.
In fact, choosing the right networking,
the performance, the throughput improvement,
going from 65% to 85% or 90%,
that kind of step up
because of your networking capability
effectively makes networking free.
Choosing the right networking,
you're basically paying,
you'll get a return on it like you can't believe
because the AI factory,
a gigawatt, as I mentioned before, could be $50 billion.
And so the ability to improve the efficiency of that factory by tens of percents results in $10, $20 billion worth of effective benefit.
And so, you know, the networking is a very important part
of it. It's the reason why NVIDIA dedicates so much in networking. It's the reason why we
purchased Mellanox five and a half years ago. And Spectrum X, as we mentioned earlier,
is now quite a, quite a sizable business. And it's only about a year and a half old.
So Spectrum X is a home run.
All three of them are going to be fantastic.
MVLink, ScaleUp, Spectrum X and InfiniBand, ScaleOut,
and then Spectrum XGS for ScaleAcross.
Your next question comes from Stacy Raskin with Bernstein Research.
Your line is open.
Hi, guys.
Thanks for taking my question.
I have a more tactical question for Collette.
So on the guy, you know, over $7 billion, the vast bulk of that is going to be from data
center.
How do I think about apportioning that $7 billion out across Blackwell versus Hopper versus networking? I mean, it looks
like Blackwell was probably $27 billion in the quarter up from maybe $23 last quarter.
You know, Hopper is still $6 or $7 billion post-H20. Do you think the Hopper strength
continues? How do I think about parsing that 7 billion
out of costs out of those three different components?
Thanks, Stacey, for the question.
First part of it, looking at our growth between Q2 and Q3,
Blackwell is still going to be the lion's share of what we have
in terms of data center.
But keep in mind that helps both our compute side
as well as it helps our networking side
because we are selling those significant systems
that are incorporating the NVLink that Jensen
just spoke about.
Selling Hopper, we are still selling it.
H100, H200s, we are.
Again, they are HTX systems and I still believe our Blackwell
will be the lion's share of what we're doing on there. So we'll continue. We don't have any
more specific details in terms of how we'll finish our quarter, but you should expect Blackwell again to be the driver of the problem.
Your next question comes from Jim Schneider of Goldman Sachs. Your line is open.
Good afternoon.
Thanks for taking my question.
You've been very clear about the reasoning model opportunity that you see,
and you've also been relatively clear about the technical specs for Rubin, but maybe you could provide a little bit of context about how you view
the Rubin product transition going forward. What incremental capability does that offer
to customers, and would you say that Rubin is a bigger, smaller, or similar step up in terms
of performance from a capability perspective relative to what we saw at Blackwell? Thank you.
Thanks.
Ruben, we're on an annual cycle.
We've been, we're on an annual cycle.
And the reason why we're on an annual cycle
is because we can do so to accelerate the cost reduction
and maximize the revenue generation for our customers.
When we increase the PERF per watt,
the token generation per amount of usage of energy,
we are effectively driving the revenues of our customers.
The PERF per watt of Blackwell will be, for reasoning systems, an order of magnitude higher than Hopper.
And so for the same amount of energy,
and everybody's data center is energy limited by definition,
for any data center that we using Blackwell,
you'll be able to maximize your revenues
compared to anything we've done in the past,
compared to anything in the world today.
And because the perf per dollar,
the performance is so good
that the perf per dollar invested in the capital would also allow you to improve your gross margins.
To the extent that we have great ideas for every single generation, we could improve the revenue generation, improve the AI capability, improve the margins of our customers by releasing new architectures.
And so we advise our partners, our customers to pace themselves and to build these data centers on an annual rhythm.
And Ruben is going to have a whole bunch of new ideas.
I paused for a second because, you know,
I've got plenty of time between now and a year from now
to tell you about all the breakthroughs that Ruben's are going to bring.
But Ruben has a lot of great ideas.
I'm anxious to tell you, but I can't right now.
And I'll save it for GTC to tell you more and more about it.
But nonetheless, for the next year,
we're ramping really hard into now Grace Blackwell, GB200,
and then now Blackwell Ultra, GB300.
We're ramping really hard into data centers.
This year is obviously a record-breaking year.
I expect next year to be a record-breaking year. I expect next year to be a record-breaking year.
And while we continue to increase the performance of AI capabilities
as we race towards artificial superintelligence on the one hand
and continue to increase the revenue generation capabilities of our hyperscalers on the other hand.
Your final question comes from Timothy Arcuri with UBS. Your line is open.
Thanks a lot. Jensen, I wanted to ask you, just to answer the question, you threw out a number,
you said 50% CAGR for the AI market.
So I'm wondering how much visibility that you have into next year.
Is that kind of a reasonable bogey in terms of how much your data center revenue should grow next year?
I would think you'll grow at least in line with that CAGR.
And maybe are there any puts and takes to that? Thanks.
Well, I think that the best way to look at it is we have reasonable forecasts from our large customers for next year.
Very, very significant forecasts.
And we still have a lot of businesses that we're still winning and a lot of startups that are still being created.
and a lot of startups that are still being created.
Don't forget that the number of startups
for AI native startups was 100 billion was funded last year.
This year, the year is not even over yet.
It's 180 billion funded.
If you look at AI native,
the top AI native startups that are generating revenues, last year was $2 billion.
This year is $20 billion.
Next year, being 10 times higher than this year is not inconceivable.
And the open source models is now opening up large enterprises,
SaaS companies, industrial companies, robotics companies to now join the AI revolution, another source of growth.
And whether it's AI natives or enterprise SaaS
or industrial AI or startups,
we're just seeing just enormous amount of interest in AI and demand for AI.
Right now, the buzz is, I'm sure all of you know about the buzz out there,
the buzz is everything sold out. H100 sold out. H200s are sold out. Large CSPs are coming out,
Large CSPs are coming out, renting capacity from other CSPs.
And so the AI native startups are really scrambling to get capacity so that they could train their reasoning models.
And so the demand is really, really high.
But the long-term outlook between where we are today, CapEx has doubled in two years. It is now running about
$600 billion a year just in the large hyperscalers. For us to grow into that $600 billion a year,
representing a significant part of that CapEx, isn't unreasonable. And so I think the next several years,
surely through the decade,
we see just really fast-growing,
really significant growth opportunities ahead.
Let me conclude with this.
BlackBell is the next-generation AI platform
the world's been waiting for.
It delivers an exceptional generational leap. Blackwell is the next generation AI platform the world's been waiting for.
It delivers an exceptional generational leap.
NVIDIA's NVLink 72 rack-scale computing is revolutionary,
arriving just in time as reasoning AI models drive order of magnitude increases in training and inference performance requirements.
Blackwell Ultra is ramping at full speed
and the demand is extraordinary.
Our next platform, Rubin, is already in fab.
We have six new chips that represents the Rubin platform.
They've all taped out the TSMC.
Rubin will be our third generation
MV-Link rack-scale AI supercomputer.
And so we expect to have a much more mature and fully scaled up supply chain. Blackwell and Rubin AI Factory platforms will be scaling into the three
to four trillion dollar global AI factory build out through the end of the decade. Customers are building ever greater scale AI factories from thousands of
Hopper GPUs in tens of megawatt data centers and soon we'll be building millions of Rubin GPU platforms
powering multi-gigawatt, multi-site AI super factories.
With each generation, demand only grows.
One-shot chatbots have evolved into reasoning agentic AI
that research, plan, and use tools, driving orders
of magnitude jump in compute for both training and inference. Agentic AI is reaching maturity
and has opened the enterprise market to build domain and company-specific AI agents for enterprise workflows, products, and services. The age of physical AI
has arrived, unlocking entirely new industries in robotics, industrial automation. Every
industry, every industrial company will need to build two factories, one to build the machines and another to build their robotic AI.
This quarter, NVIDIA reached record revenues and an extraordinary milestone in our journey.
The opportunity ahead is immense. A new industrial revolution has started. The AI race is on. Thanks
for joining us today and I look forward to
addressing you next week. Next earnings call. Thank you.
Oh, did he just release a surprise event?
This concludes today's conference call. You may not.
All right. I appreciate all of you for listening to that. How many are still here? 1,300. We've
still got a bunch of you. NVIDIA stocks still down about 2%. I thought there was a lot of really interesting headlines from that one. The China conversation, they just
said $2 to $5 billion of what they can go. No problem this quarter if the restrictions and more
could even be coming from it. I think that was a big question. Before I keep going through this
and read through some of the headlines that I was posting, I posted a bunch. Shout out to the live earnings transcript from Perplexity.
But, Montev, can I throw it over to you first and hear what your initial reactions are?
Yeah, more important than the number you just called out, they said for them, it's a 50 billion a year opportunity still, China.
So it's not something that they are going to just walk away from. So if that means
they have to whatever pay something to these guys to, you know, to let them sell, it's going to
happen. It's it's just a matter of time. We're probably Q4 at this point, I guess.
But this is not something they're walking away from.
It's a $50 billion a year opportunity growing at 50% and they expect this to...
And he went out at pains to point out, you know,
there's a massive ecosystem of AI there with, you know,
the right talent, the right companies, funding, all of that is in place. So it's not something
that they can overlook and they're going to continue to keep pushing to, you know, to get
that released. Where and what they sell there is a whole different story
it remains to be seen again whether we have uh you know we where when we get clarity on
you know the the the second uh work stop that they issued on on on the h20s that they were making
again and then also you know who is buying these you know outside
of China and what is the margin on that. The second part I want to talk about is the significance of
Harper revenues still big so between that and the complete lack of sales to China, right, that two to five billion, the
CFO in another part of the remarks said it's four to five billion.
So let's call it two to five billion accounts for the missed growth and then some without
that it would have been a significant beat even over, you know, the consensus numbers,
which is already about their their own numbers so that's
second the third thing and i really and i i posted a tweet on this i i really hope we put this to
rest for once and for all this complete nonsense about calling sales to singapore billing through
singapore as something fraudulent and and and somehow it's all being sent to China.
I have lived in Singapore. I've worked in Singapore. Every major US company has some of their
largest data centers in Singapore. Every major US company has their Asia Pacific headquarters in
Singapore. That is how they build all of their revenue to the rest of the world from
they are critical Singapore is critical to how American
companies do business with the rest of the world. There
is no big secret in this. There's no fraud in this. The
CFO said today that 99% of the revenue built through
Singapore is for US
customers.
I think we really need to put that to rest.
My friend pointed out that in another forum either I missed it somehow but even in the
last quarter she pointed this out but we still continue to hear this in spaces that somehow
this is a fraud and there's some big deal going on here. There is no deal. It's
your Google's and Amazon's and Microsoft's buying most of
this shit for their business in Singapore. That's it. Just
just put this damn thing to rest. The other thing he talked
about that's really important is they're sticking with their
annual cadence and they're talking to their customers to
build in that annual cadence so they don't have to buy the oldest or commit to buying
the older generation GPU when they can plan around the availability of the newest.
So every new data center they build will have the newest fastest technology
cheapest to run lowest power per unit of compute all of that one more thing he said they have
extremely good visibility from their large customers which is most of their business even now
and and they're pretty solid about you know what, what they see, you know, going forward in
the business and that I read it to mean that, you know, they're not expecting a sudden,
you know, disastrous drop off in business.
So at least that's how I read it.
I think those were off the top of my head, the comments I wanted to, you know, catch
and, you know, retell in my own words.
Honestly, there's a lot of interesting comments.
I think you went through a bunch of them.
One thing that wasn't noted on the call that was in the 10Q as well, or the 10Q, sorry,
was the talks around a customer customer one customer was 23%
of their total revenue a second customer was 16% so two so like 45% of their revenue for the quarter
was from two companies I wonder who they are I mean there's only three or four answers but I
wonder which one isn't them.
Yeah.
Wolfie, let's go over.
I wonder if you had any initial thoughts outside of, you know,
what Monitib was saying there, any of the points that stood out.
You should go to
someone else right now. I'm in the middle of something.
But, yeah, I apologize.
No worries.
No worries. Kevin?
I wonder if I could bring it over to you
oh he just left as well what a time what a time to be alive honestly i have family who is going
somewhere anywhere so i had a tune off at some point we'll dig in more uh in the next in the
next spaces i thought the sovereign ai conversation was interesting They said they're on track to achieve $20 billion
in sovereign AI revenue this year.
The China conversation you brought up
was really interesting.
I thought the conversation,
the one extra line he made after it,
that was you would expect it to grow
at, say, 50% per year.
But not only is that $50 billion opportunity there
or whatever, he expected it to be $50 billion opportunity there or whatever, he expects it
to be $75 billion opportunity next year in China. And that $4 trillion number of AI infrastructure
spend by the end of the decade was a number that they throwed around a couple times in the call,
which I thought they were trying to kind of address Tam a little bit there. But yeah,
I think we'll see tomorrow.
Emp did put the evil on me a little earlier by saying the last couple earnings,
NVIDIA's had big moves up,
and then kind of throughout the next day,
gave it all back.
Maybe today we had a little move down,
and tomorrow we give it back,
and we closed green.
Maybe closed green up 2-3%.
I don't know.
We'll see.
But, Gav, I see you joining us up up here we were just kind of doing a last take maybe gonna gonna
close it off here i don't know if you got the chance to listen to the call too deeply
i thought it was a good call even though the stock has not moved that i don't know how you
can walk away from that call not extremely bullish but i mean it's not a bad thing necessarily right
for the stock to stay where it was you know based on that additional
move that we first had i'm not too worried with where it ended up coming out i i i wasn't able
to follow everything within the call but uh you did mention it i was able to use um perplexity
finance to just basically summarize those earnings up for me and the call as well honestly for me to
get my post half of them i literally just was using perplexity finance there and like rewording the transcript a little bit but that was actually
really useful yeah the transcript's great the analysis is great uh like you said very bullish
on blackwell blackwell ultra those are the things that i think people are looking forward to and
seeing where they can go china i feel like they really they outline china as an opportunity more so than an issue which is i think
i guess the right way to look at it right looking for to see hey two to five billion in h
uh 20 revenue uh could be shipped in q3 so yeah i like the way that they uh propose it right there
and if they can just make some breakthroughs that'll be great ruben coming along nicely as
well i know monid have touched on that too so all the points I agree with you seem bullish. Still seems like
they're ahead of the competition. One note real quick that I would just throw in. And this was
from, I commented on John Tinsman's posts. He posted about Nvidia earnings release and the
quarterly revenue trend, basically slowing from 69% growth year
over year to 55.6% growth year over year. And I asked him what his thoughts on it, you know,
was he happy, neutral, bearish? And he said, it's the same trend that we saw in AMD last quarter.
And to him, the Nvidia earnings release is neutral, but seeing the gaming segment pickup
should help AMD. So he actually said that he saw this not only as maybe neutral for NVIDIA,
but bullish for AMD for their next earnings release.
Yeah, no, I appreciate that commentary at the end.
It's actually very interesting.
I'm looking forward to talking more about this tomorrow
on the Spaces.
Monitiv, I hope you'll come join us, Wolfie, as well as Sniper.
I'm looking forward to digesting this more. I'm sure there'll be other conversations going on,
maybe some rowdy late night spaces or something. If anyone tells you these NVIDIA earnings were
not amazing and the conversation is anything but, okay, look at where we're going to evaluate,
like look at where the expectations are coming into this. It's a $4.5 trillion company.
Anything except for that, I don't accept.
They are a supply constraint company, so missing on data center revenue.
What is that?
And that doesn't tell you as much about NVIDIA and the story we're all going here.
So I want someone to be in those late night spaces.
Maybe it's Monitiv defending uh nvidia if anyone wants
to come in and and make some claims they're clearly not there so i appreciate everyone
you should follow speakers if you guys have any words extra kind of things in there i don't have
a hard hard cut off here but yeah i just want to add they are they are a regulation constrained
company yeah a hundred percent They are a supply and regulation
constraint in different areas. Imagine if you take
everything away and they're still doing this
revenue? Are you kidding me?
Interesting
times. Follow the speakers.
This is a great space. The whole thing was recorded.
You can go back and listen.
Stocks on Spaces, we're doing this
type of content
every single Monday through Thursday, 3 to 5 p.m. Eastern, at least.
It's 6, 10 p.m. Eastern here, so going a little after extra there.
You should also be following that Wolf account that is up here
and the Wolfie account.
You should follow everyone that's up here, but shout-out to that Wolf account
if you enjoy these spaces.
A lot more are coming out from there.
But, yeah, I appreciate everyone.
Have a great day, team.
We'll catch you all later.
We are going to dig in a lot, a lot more.
We're going to dig in a lot tomorrow.
Shout out to Emp.
We've missed you on the spaces, by the way, too.
But we're going to dig in a lot more
on these NVIDIA numbers and everything tomorrow
and have plenty of time.
So I appreciate you all.
We'll miss you, Emp.
And have a great day team
thank you for your attention to this matter
thank you for your attention
I'm going to go have some bad time Thank you.
