Broadcom reported a strong Q1, with revenue topping estimates by $330 million as management noted that AI revenue rose 77% YoY in the quarter. AI revenue reached $4.1 billion this quarter, beating previous guidance of $3.8 billion. The beat was driven by strength in networking with management mentioning last quarter that custom silicon would see stronger growth in H2. It was also stated networking was 40% of AI revenue while it’s typically 30% of AI revenue (confirming the higher product mix this quarter). 

The stock was up double-digits after hours specifically when Broadcom stated next quarter AI revenue would be $4.4 billion and when management reiterated their serviceable addressable market forecast of $60 billion to $90 billion, stating “these R&D investments are very aligned with the roadmap of our three hyperscale customers as they each race towards 1 million XPU clusters by the end of 2027. And accordingly, we do reaffirm what we said last quarter, that we expect these three hyperscale customers will generate a Serviceable Addressable Market or SAM in the range of $60 billion to $90 billion in fiscal 2027.”  

Managment also mentioned there are two more hyperscalers that plan “to tape out their XPUs” this year, referencing future expansion of the quoted SAM.  

We’ve seen many AI stocks with similarly strong reports sell off this quarter, whereas the market loves Broadcom’s strong margins, its cash and the clarity management provides in terms of exact AI revenue per quarter.  

Revenue  

Broadcom’s revenue increased 24.7% YoY to $14.92 billion in its fiscal Q1, ahead of estimates for $14.59 billion due to strong AI momentum. Management guided for Q2 revenue to be flat sequentially at $14.9 billion, ahead of estimates for $14.73 billion.  

Looking ahead, growth is expected to decelerate to 19.3% in Q2, a more than 5 point deceleration from 24.7% growth in Q1. Broadcom is lapping comps that include VMWare’s acquisition, hence why there is a large QoQ deceleration in Q1.  

Most importantly, AI revenue rose 77% YoY and ~11% QoQ to $4.1 billion. For Q2, Broadcom said that it expects AI momentum to continue in Q2 “as hyperscale partners continue to invest in AI XPUs and connectivity solutions for AI data centers,” forecasting AI revenue to be $4.4 billion, up just over 7% QoQ and 44% YoY. This would put AI revenue at $8.5 billion for the first half of the fiscal year, up more than 57% from $5.3 billion in the same period last year. 

Key Segments 

Semiconductor Solutions revenue rose 11% YoY and was approximately flat QoQ at $8.22 billion, decelerating 1 point from 12% YoY growth in Q4.  

• AI revenue was $4.1 billion, ahead of management’s guidance for $3.8 billion on “stronger shipments of networking solutions to hyperscalers on AI.” Management added that they “see a steady ramp in deployment of our XPUs and networking products,” supporting its AI revenue guide of $4.4 billion in Q2. 

Non-AI semiconductor revenue was $4.1 billion, down 9% QoQ due to a seasonal decline in wireless revenue.  

• Broadband “showed a double-digit sequential recovery in Q1 and is expected to be up similarly in Q2 as service providers and telcos step up spending.” 
• Server storage “was down single-digits sequentially in Q1, but is expected to be up high-single digits sequentially in Q2.” 
• Enterprise networking is expected to remain flattish in Q1 and Q2 as customers work through inventory. 
• Wireless was flat in Q1 and expected to be flat YoY in Q2, and resales in industrial were “down double-digits in Q1 and are expected to be down in Q2.” 

Infrastructure Solutions revenue rose 47% YoY and 15% QoQ to $6.70 billion. Management said the QoQ performance was “exaggerated though by deals which slipped” from Q4 to Q1, though the strong YoY performance was driven by two factors – converting largely perpetual licenses to one full subscription, and “upselling customers to a full stack VCF, which enables the entire data center to be virtualized.” At the end of Q1, approximately 70% of Broadcom’s 10,000 largest customers have adopted VCF.  

• For Q2, Infrastructure Software revenue is expected to be $6.5 billion, up 23% YoY. 

Margins 

Margins expanded significantly across the board in Q1, with GAAP operating margin seeing the strongest expansion on a QoQ basis of 9.1 points.  

• Gross margin was 68.0% in Q1, up from 64.1% in the prior quarter as gross profit topped $10 billion for the first time. Adjusted gross margin was 79.1%, expanding from 76.9% in the prior quarter. 
• Operating margin witnessed the strongest QoQ expansion, with Broadcom reporting a 42.0% margin in Q1, up from 32.9% in Q4; this is marking a return to pre-VMWare acquisition levels. Adjusted operating margin was 65.9%, up from 62.7% in the prior quarter.  
• Net margin was 36.9% in Q1, a strong 6.1 point expansion from 30.8% in the prior quarter. Adjusted net margin was 52.4%, up from 49.6% in the prior quarter. 

Adjusted EBITDA margin also expanded nearly 3 points sequentially to 67.6%, with adjusted EBITDA surpassing $10 billion for the first time in Q1 at $10.08 billion. This also marked a strong, consistent expansion from the high-59% range at the start of FY24 when VMWare’s integration was impacting margins. Q2’s adjusted EBITDA margin was guided at 66%, a slight sequential contraction. 

EPS 

Given the strong margin expansion in Q1, Broadcom delivered a strong GAAP EPS beat of 35.7%, while its adjusted EPS beat was smaller in nature at ~6%. 

• GAAP EPS of $1.14 beat estimates for $0.84.  
• Adjusted EPS of $1.60 beat estimates for $1.51, for growth of 45.5% YoY. This is expected to be peak growth for adjusted EPS in fiscal 2025, with estimates pointing to 37% growth in Q2 below ending the fiscal year at almost 23% growth. 

Cash and Balance Sheet 

Operating and free cash flow were both strong in Q1 due to the margin strength and revenue outperformance, with both recording a margin of >40%.  

• Operating cash flow was $6.11 billion, up 9% QoQ. OCF margin was 41%, improving from 39.9% last quarter. 
• Free cash flow was $6.01 billion, up almost 10% QoQ. FCF margin was 40.3%, improving from 39.0% last quarter. 
• Inventories were $1.91 billion, increasing more than 8% QoQ. 
• Cash and equivalents totaled $9.31 billion, while debt decreased $1 billion QoQ to $66.58 billion. 

Earnings Call: 

Market Opportunity: 

The market opportunity for Broadcom is quite large – represented by SAM of $75B at the midpoint by 2028, up from a $16B run rate right now. An analyst asked a similar question in terms of chips and management reiterated their current forecast is with only three customers right now whereas two more are likely to go into volume production in the coming years. 

Timothy Arcuri: 

Thanks a lot. Hock, in the past, you have mentioned XPU units growing from about 2 million last year to about 7 million you said in the 2027, 2028 timeframe. My question is, do these four new customers, do they add to that 7 million unit number? I know in the past, you've sort of talked about an ASP of 20 grand by then. So those — the first three customers are clearly a subset of that 7 million units. So do these new four engagements drive that 7 higher, or do they just fill in to get to that 7 million? Thanks. 

Hock Tan: 

And thanks, Tim for asking that. To clarify, as I made — I thought I made it clear in my comments. No, the market we are talking about, including — when you translate the unit is only among the three customers we have today. The other four, we talk about engagement partners. We don't consider that as customers yet and, therefore, are not in a served available market. 

China Exposure: 

Management was asked how many of the hyperscalers were from China, but they declined to comment. This is a notable question given any hyperscaler from that region has a low chance of being unscathed over the next four years. There were reports earlier this week that Broadcom may be losing major customer ByteDance, for example. 

Management stated “no comment” yet it’s important to note the concern. 

Conclusion: 

Broadcom is quality and will help us hedge any anti-Nvidia narratives – which are bound to come up from time to time. As far as quality goes, you cannot find a better pair in AI (NVDA-AVGO). These are the juggernauts and we want exposure to AVGO as its empire slowly expands. Don’t forget AVGO’s AI software opportunity, which we’ve expanded on in the past. 

That’s officially a wrap for the I/O Fund earnings season!  

Regarding the AI trade falling out of favor this quarter, don’t let the market fool you for one minute – investors have never had it so good as to be able to track and verify there are hundreds of billionshundreds of billions pouring into one trend. It’ll be volatile, and it’ll be scary – and then ultimately thrilling, but this trend is bursting at the seams with demand. Supply is not merely bottlenecked; it’s dammed at the flood gates. The supply issues will work themselves out, and the I/O Fund and our Members will be well-positioned when those flood gates open.

Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.

Recommended Reading:

• Marvell Q4: Margin Expansion Yet AI Capitulation
• Nvidia Q4: Range Bound and Looking for a Catalyst
• AI Hardware Suppliers Forecast Muted H1, Strong H2 — and What it Might Mean for Nvidia
• AMD Q4 Earnings: Q1 Decline in Data Center, Mgmt Drops AI Guidance

Marvell reported in line results with Q4 revenue that grew by 27.4% YoY and 19.9% QoQ to $1.82 billion, beating estimates by 1.2%. Adjusted EPS grew by 30.4% YoY to $0.60, beating estimates by 1.6%. 

Management guided Q1 revenue of $1.875 billion, representing a YoY growth of 61.5% YoY and 3.2% QoQ at the midpoint, beating estimates modestly by 0.3% and expects adjusted EPS to grow 154.2% YoY to $0.61, beating estimates by 1.7%. 

The company’s margins have improved from negative GAAP operating margin (often double-digit negative) to 12.9% in the most recent report. The adjusted operating margin of 33.7% was flat but on higher revenue, resulting in adjusted EPS that grew 40% sequentially.  

The market is going through a phase of AI capitulation. Our site is particularly helpful here as Knox sets out price targets months in advance on his webinar for AI stocks, offering two scenarios – a pullback scenario and a breakout scenario. The AI selloff is behaving according to the scenarios he set forth months ago.  

But for now, it’s retail against the Street. The Street is confusing narratives, lowering price targets despite blowout earnings (those lowered price targets are 100% above the price in some cases), and more specifically to Marvell, stating Amazon Trainium revenue should have been higher. Management stated AI revenue was substantially above the $1.5 billion target for FY2025, and they expect to exceed the $2.5 billion target in FY2026 significantly. The overall revenue growth accelerated in the second half of the year, driven by the custom silicon program ramps along with continued strong growth in Electro-Optics. However, these numbers have not changed for a few quarters now (the above $1.5B and $2.5B) and I’m sure the Street wanted more in terms of a new number given the 5-year deal with Amazon, implying Trainium2. 

Although Marvell’s quarter was not a blowout, with some readthroughs being the Amazon deal is not as strong as expected, inventory is up 20% QoQ compared to 3% QoQ revenue guide, and management stated they are in “significant volume production ahead.” Overall, I’ve seen supply taking longer this quarter, thus within the framework of what I’ve seen, Q4 reports and Q1 guides are an air pocket of sorts across the board. Ultimately, there is $300B in capex saying the air pocket will resolve in the (patient) AI investor’s favor. 

Hopefully, the multi-dimensional approach we offer by combining fundamentals and technicals – not only for lower entries that are discussed months in advance but also going so far as to hedge up to 100% of our portfolio — has proven its value over the past few months. Marvell is the most over-sold its been in its history – that is what is meant by capitulation. A bounce in price is on the horizon. 

Revenue growth Headed to 61.5% Next Quarter

The company’s Q4 revenue grew by 27.4% YoY and 19.9% QoQ to $1.82 billion, driven by strong AI demand, beating estimates by 1.2%.  

• Management guided Q1 revenue of $1.875 billion, representing a YoY growth of 61.5% YoY and 3.2% QoQ at the midpoint, beating estimates modestly by 0.3% 
• Analysts expect revenue to grow 55.9% YoY to $1.98 billion in Q2 and 39.1% YoY to $2.11 billion in Q3. 

• FY2025 revenue grew by 4.7% YoY to $5.77 billion. 
• Analysts expect FY2026 revenue to grow 42.3% YoY to $8.21 billion and 20.5% YoY to $9.89 billion in FY2027.

Margins – GAAP Profitable

The company exceeded its margin guidance in Q4, driven by operating leverage, and achieved GAAP profitability. Management expects GAAP profitability to continue during FY2026.  

• Q4 gross margin was 50.5% compared to 46.6% in the same period last year. Gross margin was slightly above the guidance of 50%. Management guide for Q1 is 50.5%.  
• Adjusted gross margin was 60.1% compared to 63.9% in the same period last year. Management guide for Q1 is 60%. 
• Q4 operating margin was 12.9% compared to (-2.3%) in the same period last year. It was above the guidance of 10.6%. Management guide for Q1 is 12.5%.  
• Adjusted operating margin was 33.7% compared to 33.8% in the same period last year. Managment guide for Q1 is 34%. Management expects strong operating leverage to continue in FY2026 and expects to make significant progress towards the long-term non-GAAP operating margin target of 38% to 40%.  
• Q4 net income was $200.2 million or 11% of revenue compared to a net loss of ($392.7 million) or (-27.5%) of revenue in the same period last year. Adjusted net income was $531.4 million or 29.2% of revenue compared to $401.6 million or 28.2% of revenue in the same period last year.  
• The difference in the GAAP net income and non-GAAP net income was due to stock-based compensation of $147.6 million or 8.1% of revenue and amortization of acquired intangible assets charge of $169.5 million or 9.3% of revenue. 

Adj.EPS grew by 30.4% YoY and 39.5% QoQ 

Q4 GAAP EPS was $0.23 compared to (-$0.45) in the same period last year. Q4 adjusted EPS grew by 30.4% YoY and 39.5% QoQ to $0.60 driven by operating leverage, beating estimates by 1.6%. 

• Management expects Q1 GAAP EPS of $0.19 and adjusted EPS to grow 154.2% YoY to $0.61, beating estimates by 1.7%. 
• Analysts expect Q2 adjusted EPS to grow 121.6% YoY to $0.66 and 69.5% YoY to $0.73 in Q3. 
• Analysts expect FY2026 adjusted EPS to grow 77.5% YoY to $2.79 and 31.8% YoY to $3.67 in FY2027.  

Cash Flow and Balance Sheet

The company’s cash flow margin was lower due to the increase in inventories to support the strong expected growth.  

• Q4 operating cash flow margin was 28.3% compared to 38.3% in the same period last year.
• Q4 free cash flow margin was 24.4% compared to 32.6% in the same period last year. 
• Cash was $948.3 million and debt of $4.06 billion compared to $868.1 million and $4.1 billion at the end of Q3.  
• The company paid $52 million in dividends and repurchased shares worth $200 million in Q4. 

Management mentioned that they received an upgrade to investment grade credit rating from Fitch which is positive as it helps the company to refinance its high debt with better terms in the future. The company’s CFO Willem Meintjes said in the earnings call, “We were pleased to receive an upgrade to our investment grade credit rating from Fitch in January, citing their positive outlook on Marvell's strong operating momentum from robust data center demand, structurally improved leverage metrics, strong market position and strengthened cash flow profile.” 

• Inventories increased to $1.03 billion from $859 million in Q3, up 19.9% sequentially to support the strong expected growth. 

Key Segments 

Data Center 78% YOY and 24% QOQ

The company achieved record Q4 revenue of $1.37 billion, growing 78% YoY and 24% QoQ. The strong results were driven by the custom AI silicon programs ramping to high volume production. Additionally, the company also benefited from strong shipments of Electro-Optics products and Teralynx Ethernet switches with revenue from both product lines growing double-digits sequentially on a percentage basis. Management pointed toward the 800G PAM4 products as being the workhorse in its electro-optics products. Marvell is also at the leading edge of 1.6T PAM DSPs and will be first to introduce the 3nm 1.6T DSP with 200G per lane. 

AI Revenue Commentary 

Management mentioned that AI revenue was substantially above the $1.5 billion target for FY2025, and they expect to exceed the $2.5 billion target in fiscal 2026 significantly. The overall revenue growth accelerated in the second half of the year, driven by the custom silicon program ramps along with continued strong growth in Electro-Optics. 

Here is what was said in the opening remarks: 

“We ended the year with our AI revenue substantially above our $1.5 billion target from April 2024's AI Day and we also expect to very significantly exceed our $2.5 billion target in fiscal 2026.” It was later stated: “Last year, we had talked about $1.5 billion. We blew through that — this year, again, we anticipate being substantially above that. I'm not putting a number on it just yet. I think that there's a lot to go here in terms of the momentum in the business and the opportunity set in front of us. And so we're, right now, we're kind of keying off the last update we did, which was in the AI Day from last year and then we'll find the right appropriate time in the future.” 

Carrier Infrastructure 

Carrier Infrastructure Q4 revenue was down (-38%) YoY and up 25% QoQ to $105.8 million. Revenue accelerated from the (-73%) YoY decline and 12% sequential growth in Q3. 

Enterprise Networking  

Enterprise Networking Q4 revenue declined by (-35%) YoY and up 14% QoQ to $171.4 million. Revenue accelerated from a decline of (-44%) YoY and flat QoQ in Q3. The company witnessed continued recovery in both Carrier Infrastructure and Enterprise Networking with revenue collectively growing 18% sequentially. Management expects aggregate revenue from enterprise networking and carrier infrastructure to grow sequentially by approximately 10% in Q1. 

Consumer End Market 

Q4 Consumer End Market revenue declined by (-38%) YoY and (-8%) QoQ to $88.7 million. For Q1, due to seasonality in gaming demand to expected to drive a sequential decline of (35%) in the consumer end market. Over the next several years, management expects the consumer end market revenue to be approximately $300 million on an annual basis. 

Automotive/Industrial 

The automotive/industrial revenue grew by 4% YoY and 3% QoQ to $85.7 million in Q4. The management expects continued sequential growth in the automotive end market. However, this growth will be more than offset by a decline in revenue from the industrial end market, where order patterns can be lumpy in any given quarter. As a result, they expect the overall revenue from the auto and industrial end market to decline sequentially in the high-single-digits on a percentage basis for Q1.

China Revenue 

Marvell has a high China revenue concentration. It constituted 43% of revenue in Q3 and 45% for the nine months ending Q3. There was no mention of China during the recent earnings call, and we need to wait for the 10-K report to know the latest percentage of revenue from China. The high revenue concentration is a concern, notably since the recent tariffs increased from 10% to 20%. 

Earnings Call: 

XPU Design Discussions (Likely Amazon)

Regarding the custom design projects, Marvell stated the following in the opening remarks: “Let me now turn to our current custom silicon programs. Marvell has successfully ramped highly complex 100 billion plus transistor XPUs and CPUs from initial samples to high volume production on first pass silicon […] As I mentioned, our two leading AI custom programs are in high volume production, and we expect growth to continue. One of these is a custom ARM CPU, which we expect we'll see expanding adoption in our customers' data centers. The second program is for a custom AI XPU, which is also performing extremely well with significant volume production ahead. In parallel, we are fully engaged with this customer on the follow-on generation of this XPU and planning for a production ramp once it completes its sampling and qualification cycles. As a result, we expect our revenue from custom XPUs for this customer to not only grow this year, fiscal 2026, but continue to grow next year fiscal 2027 and beyond.” 

There is a second hyperscaler, perhaps Meta, with potential “to start production in calendar 2026.” 

In the call, analysts asked more about the lead customer. From the Q&A discussions below, I did not have a readthrough that Marvell lost Amazon’s Trainium2 business: 

Harlan Sur 

Hey, good afternoon guys. Thanks for taking my question. Great to see that you captured the follow-on AI XPU program after your current XPU program, which is ramping now with your major cloud and hyperscale customer. Matt, just to clarify. So is the new follow-on XPU program, a training XPU as well? Is that a calendar '26 ramp? And is that at 5-nanometer or 3-nanometer? Any more color there would be helpful. And then for Willem, your inventories were up 20% sequentially, which in a strong demand and product cycle environment like typically implies strong future growth, but you compare that to the 3% sequential total revenue guide for April. There seems to be some disconnect. So the way to interpret this is that the 20% sequential growth in inventories is more reflective of the AI strong growth profile ahead for the team? 

Matt Murphy 

Yes, I'll start off, Harlan. So, yes, so couple of things. So again given the confidentiality wrapper we've got, here we go. The first is you should assume this is a — it's a very high volume program and it's a continuation of what we're doing. It on the — and you should assume just in general, on every next generation type of device, you're going to see no transitions and technology advancements as you go forward. So you should assume that. And then also on a timing perspective, all I can say there is we'll be ready to ramp when it's time, and we'll manage that transition. We're very confident in our ability to manage that transition successfully with our customer, but that timing is something that we're just going to have to see when that's ready and we'll time it around that. And I can't really comment on what my customer plans are in this kind of detail, they just don't like it, and I don't blame them. And I'll give — I'll let Willem take the inventory question. Thanks. 

Note: the answer above is about the next program which management is reluctant to discuss due to confidentiality 

While the above quote was clear that another analyst was not expressing concerns about the loss of the lead customer (quite the opposite), it was perhaps even more clear on management’s side in this fairly long exchange: 

Ben Reitzes: 

Hey, guys. Thanks for the question and thanks for the intermission there in the middle. That was a nice break. The question that I have is with regard to sequential growth, Matt, and then a long-term question. Did you clarify that AI revenue could grow sequentially through the year? And did — when Tim asked his question earlier? And if so what's your confidence? And then if you could just kind of step back also, Matt, a lot of noise due to the speculation around that customer and content there. But when you step back, you have a goal of $15 billion for the data center long-term by calendar year '28 are you seeing the progress in your custom business still to hit that number because nobody on the street is even close to that. Thanks. 

Matt Murphy: 

Yes, Ben, thanks for the questions. So, yes, on the assumptions through the year, yes, I didn't, I think what I said was in Q1 our data center business ex the on-prem stuff was going to be up double-digits, coming off of overall data center like 25% a quarter sequentially. So I didn't comment throughout the whole year, but you should just you should just assume. It's not a bad assumption, right, or would be a fair assumption to think that it's obviously going to continue given the strength in the business and the momentum that we're seeing and just from what we're looking at from a year-over-year perspective. On the long-term, I think we're tracking extremely well to our 20% market share number. When we look back to kind of calendar '23 and then '24 and '25 and you start bouncing it up against what we said at the AI day. I think it looks very favorable. We're definitely gaining share from '23 to '24 and we'll definitely gain more share from '24 to '25. And so to get to that revenue target, you got to get there both ways. You got to grow the share, right, from kind of 10%, let's call it to 20%. And then the market's got to develop, right, obviously. You got to get to the sort of $75 billion TAM, but both of those are trending in a very positive direction. In fact, certainly in '24 and even in '25 is just kind of big round numbers, it looks like both the market and our growth, if you just bounce it up against sort of what's the compounded growth rate you needed from '23 to '28, it's actually growing above that right now. So now when we're in ramp phase, but that's going to continue. So I think we're in very good shape in terms of where we're tracking from a market share perspective. And certainly, if the market, by the way, is bigger. I mean remember, we gave that point of view in April of last year. And the world has changed then since then in terms of the absolute CapEx that's being deployed, certainly, even in recent months, all of the various programs from all the different key players and the potential of what's out there, sovereign programs, government programs, programs from new entrants. So we just see this — we just see the — quite frankly the TAM and the opportunity for Marvell, if anything, being way larger than it was when we looked at it almost a year ago. So all those make me feel very good about the market size developing and then certainly our progression on the market share. And you're right, that would be, it'd be an absolute home run to get there. And that's what me and my team are absolutely driving in this company day in and day out is drive the market share and help create the market, help make the TAM happen and execute like crazy and do it in a very focused manner, which is also why we reorganized the company with a dedicated data center engineering and business group to drive it. So I think the setup is really good. Thanks. 

Networking Opportunity Expands TAM 

We’ve covered Marvell’s networking products in great detail for six years now in previous analysis – this one is the most current. It’s a toss up as to whether Marvell’s custom silicon or networking opportunity will end up being the bigger market in a few years' time. 

What was most interesting was that management discussed networking being incremental TAM to what was discussed above on custom silicon.  

“The key thing I want to stress to you and to the investors on the call is this is incremental TAM the scale-up opportunity. We flagged it a few different times is right for sort of a TAM expansion. And it looks like that's what's going to play out at some point here. So that is all incremental. That's not, hey, if that happens and all of a sudden, Marvell's DSP revenue goes down or something like the scale up and this type of connectivity that is a revenue upside, market upside type of opportunity. So that one, I think, is — we're very excited about.” 

Conclusion: 

Despite design companies being in Nvidia’s shadow, Marvell put up a decent report. The company started the year at a run rate of $4 billion-ish and is now on a $7.5 billion run rate. They stated 75% of their revenue is from data center (this includes AI and on-prem, etc). Each investor will need to decide for themselves if the commentary or fundamentals reflects a weaker lead customer (Amazon). That’s not my readthrough from this report. 

However, Marvell has higher debt and higher China revenue than our other semis, and that’s why we aren’t buying the report. We prefer to build other positions with less risk in these two areas, while acknowledging Marvell remains a strong AI contender – for both custom silicon and networking. 

With that said, Marvell has never been more over-sold (technicals-wise) and is due for a bounce. Here is the note I got from Knox this morning, he will elaborate more in today’s webinar:  

“Marvell has been trading for over 25 years. In this period, we have only seen the current level of oversold conditions 3 other times – 7/9/02, 10/9/08, 8/2/10. The first 2 were toward the tail end of bear markets: the period in 2022 period was followed by a +50% rally, before turning lower, while the period in 2008 was followed by a +20% rally in a short amount of time before eventually finding a bottom. The third period was not during a bear market and led to a +50% rally over the span of 6 months, before turning lower.”

Royston Roche, Equity Analyst at the I/O Fund, contributed to this article.

Please note: The I/O Fund conducts research and draws conclusions for the Fund’s positions. We then share that information with our readers. This is not a guarantee of a stock’s performance. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis.

Recommended Reading:

• Credo Fiscal Q3: Hypergrowth, Strong Margins, and AI Accelerator Agnostic
• Dell Q4: Projects $15 billion in AI shipments this year
• Nvidia Q4: Range Bound and Looking for a Catalyst
• Nvidia Q4 Financials: Beat Likely for Q4; Questions Persist for Q1
• AI Hardware Suppliers Forecast Muted H1, Strong H2 — and What it Might Mean for Nvidia

Compared to any other stock on Wall Street, Nvidia reported an exceptional quarter. Yet, Nvidia’s stock is competing with itself at this point, and when compared to previous Nvidia earnings reports, Q4 and Q1 represent a pause in the fireworks. 

Nvidia reported $11 billion in Blackwell revenue, however, Huang finally admitted the NVL systems are in production (i.e., not shipping in volume). Rather, other SKUs are making up Blackwell revenue for now. According to a discussion with an analyst, Nvidia foresees NVL systems and the next generation Blackwell Ultra ramping “simultaneously” — which increases the pressure to deliver this fiscal year given there was a “couple months” delay with the NVL systems.  

Fortunately, for the I/O Fund, we use technicals to reduce risk. When there is a change in story, even if it’s transitory by a quarter or two, we rely on technicals to protect our position while maximizing the upside. As you know, we are exposed to Nvidia — yet heavily exposed when you consider its ecosystem. 

We also hold our management teams accountable, and not even Jensen Huang should get a free pass. Investors should have been provided more transparency on when the premier SKU will be shipping in volume, since it was originally slated for Q1. Below, I outline the reasons that H2 is likely to be the catalyst for the next leg higher, whereas Q1 may not have been enough to cause the stock to break out. With that said, semiconductors see intra-quarter news more frequently than software peers, and thus, this could change anytime.  

Nvidia’s valuation remains very attractive. I stated on Fox yesterday that Nvidia and its suppliers are the best trade of the year – the question that remains is timing. If the NVL systems were shipping in volume, the supply chain would be on fire right now – is what I mean by timing. I also pointed out that investors do better when proxies participate – whether that’s suppliers providing a clear, green light or an ETF like SMH, which is 14% off ATHs. 

I look at this and more below! 

Revenue Beat/Raise (Again) 

Nvidia beat consensus estimates by just $1.15 billion this quarter, its smallest beat of the last seven quarters, as it reported revenue of $39.33 billion, up 77.9% YoY and 12.1% QoQ. Management reported that Blackwell was witnessing the fastest product ramp in company history, delivering $11 billion in Blackwell revenue in Q4. 

For FY25, Nvidia reported revenue growth of 114% YoY to $130.5 billion, marking a consecutive year of triple digit growth after revenue rose 126% YoY in FY24. 

For Q1, Nvidia guided for revenue of $43 billion, +/- 2%, representing growth of 65.1% YoY and 9.3% QoQ at midpoint, slightly ahead of estimates for 61.2% YoY growth to $42 billion. As we had noted in our pre-earnings analysis, some analysts were looking for as much as $47 billion in revenue in Q1. On a sequential basis, Q1 is also estimated to see QoQ growth dip to the single-digit range at the midpoint of guidance, slowing nearly 3 points from Q4.  

Margins to Improve H2 

Margins are expected to contract sequentially in Q1 as Blackwell ramps, with management having stated in Q3 that gross margins were expected to dip to the low-70% level as Blackwell begins to ramp before returning to the mid-70% range. 

Gross margins in Q4 were in-line with prior guidance, though operating margins came in ahead of guidance as Nvidia continued to benefit from strong operating leverage. 

• Q4 GAAP gross margin was 73%, contracting from 74.6% in Q3. For Q1, GAAP gross margin is expected to contract further to 70.6%.  
• Adjusted gross margin was 73.5% in Q4, down from 75% in Q3, with management guiding for another contraction to 71% in Q1. 
• Q4 GAAP operating margin was 61.1%, contracting from 62.3% in Q3. For Q1, GAAP operating margin is expected to drop below 60%, with management’s expense guidance implying a 58.5% margin.  
• Adjusted operating margin was 64.9% in Q4, down from 66.3% in Q3, with Q1 forecast to contract further to 62.6%. 
• Q4 GAAP net margin expanded slightly sequentially to 56.2%, though Q1’s net margin is implied to be 50.1% due to the QoQ gross and operating margin contractions. 

For FY25, margins expanded on a YoY basis despite contracting in the back half of the year. FY25’s GAAP gross margin was 75%, up from 72.7% in FY24. GAAP operating margin expanded 8.3 points YoY to 62.4%, as GAAP operating income increased 147% YoY to $81.5 billion. GAAP net margin increased 7 points YoY to 55.9% with a similar 134% YoY rise in net income to $86.8 billion. 

EPS 

Q4 GAAP EPS increased 82% YoY to $0.89, beating estimates for $0.80. Adjusted EPS of $0.89 beat estimates for $0.85. Growth is expected to moderate to the ~50% range in the coming quarters, likely pressured by falling margins due to Blackwell’s ramp. 

• Q1 adjusted EPS is expected to rise 49.4% YoY to $0.91. 
• Q2 adjusted EPS is expected to rise 50.2% YoY to $1.02. 

Cash Flows and Balance Sheet 

Operating and free cash flows both decreased slightly sequentially, with management explaining that this was due to a “a higher accounts receivable balance due to shipment linearity and increased inventory to support our Blackwell product ramp.” Both inventories and accounts receivables showed strong QoQ growth. 

• Operating cash flow in Q4 was $16.63 billion, down from $17.63 billion in Q3. OCF margin was 42.2%, more than 8 points lower than Q3’s 50.3% margin. FY25’s operating cash flow was $64.09 billion, rising 128% YoY and representing a margin of 49.1%, up 3 points YoY. 
• Free cash flow was $15.52 billion in Q4, down from $16.79 billion in Q3. FCF margin was 39.5% in Q4, down from 47.9% in Q3. For FY25, free cash flow rose 125% YoY to $60.72 billion, for a margin of 46.5%, up 2.3 points YoY. 
• Cash and marketable securities totaled $43.21 billion, while debt totaled $8.46 billion. 

Inventories, accounts receivable and purchase commitments all rose quite substantially on a sequential basis as Nvidia prepares to ramp Blackwell throughout the year. 

• Inventories rose nearly 32% QoQ to $10.08 billion, representing a sequential rise of $2.43 billion.  
• Accounts receivable rose $5.4 billion QoQ to $23.1 billion. 
• Purchase commitments and obligations for inventory and manufacturing capacity rose $1.9 billion QoQ to $30.8 billion. 

Key Segments 

Data Center 

Data center revenue rose 93% YoY and 16% QoQ to $35.58 billion in Q4, driven primarily by GPUs as networking revenue was soft. Compute revenue rose 116% YoY and 18% QoQ to $32.56 billion in Q4, though Networking revenue declined (9%) YoY and (3%) QoQ to $3.02 billion. 

Management noted that they “delivered $11.0 billion of Blackwell architecture revenue in the fourth quarter of fiscal 2025, the fastest product ramp in our company’s history,” driven by major CSPs which accounted for 50% of data center revenue. Q4 growth was also aided by sequential growth for the H200. 

For Networking, Nvidia said that it is “transitioning from small NVLink 8 with Infiniband to large NVLink 72 with Spectrum X,” with growth in Ethernet and NVLink products related to the Grace Blackwell ramp. 

For FY25, data center revenue rose 142% YoY to $115.2 billion, driven by a 162% YoY increase in Compute revenue to $102.2 billion, with Networking revenue rose just 51% YoY to $12.99 billion. 

Other Segments 

• Gaming revenue declined (11%) YoY and (22%) QoQ to $2.54 billion, due to “limited supply for both Blackwell and Ada GPUs.” For FY25, gaming revenue rose 9% YoY to $11.35 billion. 
• Automotive revenue rose 103% YoY and 27% QoQ to $570 million, driven by self-driving platform sales. For FY25, automotive revenue rose 55% YoY to $1.69 billion. 
• Pro Viz revenue rose 10% YoY and 5% QoQ to $511 million, driven by a “continued ramp of Ada RTX GPU workstations for use cases such as generative AI-powered design, simulation, and engineering.” For FY25, Pro Viz revenue rose 21% YoY to $1.88 billion. 
• OEM and other revenue rose 40% YoY and 30% QoQ to $126 million. For FY25, OEM and other revenue rose 27% to $389 million.

Earnings Call:

NVL 72 Systems and Blackwell Ultra 

Of the $200 billion in revenue expected this year, roughly $100 billion is expected to come from NVL systems which comes out to 30,000 to 35,000 racks at $3 million per rack. Given there is evidence of a delay, there was bound to be a question about this on the call. Overall, the answer helps to alleviate if the systems are ready yet (they have “ramped production” but not stated they were shipping in volume).  

The remaining issue is if both Blackwell’s NVL systems and Blackwell Ultra (the next generation of GPUs) can ramp "simultaneously” (which was not the original road map). Perhaps these two can successfully ramp simultaneously, but it’s also important to acknowledge this was not the original plan. Personally, I think the hiccup discussed below should be explained more as to when the systems will ship in volume, since it was originally expected to be Q1. Judging by analyst estimates, the October quarter will see higher QoQ growth in terms of revenue at $5 billion compared to analyst estimates of $4 billion for the previous quarter.  

Harlan Sur  

Good afternoon. Thanks for taking my question. Your next-generation Blackwell Ultra is set to launch in the second half of this year, in line with the team's annual product cadence. Jensen, can you help us understand the demand dynamics for Ultra given that you'll still be ramping the current generation Blackwell solutions? How do your customers and the supply chain also manage the simultaneous ramps of these two products? And — is the team still on track to execute Blackwell Ultra in the second half of this year? 

Jensen Huang  

Yes. Blackwell Ultra is second half. As you know, the first Blackwell was we had a hiccup that probably cost us a couple of months. We're fully recovered, of course. The team did an amazing job recovering and all of our supply chain partners and just so many people helped us recover at the speed of light. And so now we've successfully ramped production of Blackwell. 

But that doesn't stop the next train. The next train is on an annual rhythm and Blackwell Ultra with new networking, new memories and of course, new processors, and all of that is coming online. We've have been working with all of our partners and customers, laying this out. They have all of the necessary information, and we'll work with everybody to do the proper transition. This time between Blackwell and Blackwell Ultra, the system architecture is exactly the same. It's a lot harder going from Hopper to Blackwell because we went from an NVLink 8 system to a NVLink 72-based system. So the chassis, the architecture of the system, the hardware, the power delivery, all of that had to change. This was quite a challenging transition. 

But the next transition will slot right in Blackwell Ultra will slot right in. We've also already revealed and been working very closely with all of our partners on the click after that. And the click after that is called Vera Rubin and all of our partners are getting up to speed on the transition of that and so preparing for that transition. And again, we're going to provide a big, huge step-up. And so come to GTC, and I'll talk to you about Blackwell Ultra, Vera Rubin and then show you what we place after that. Really exciting new products, so to come to GTC piece. 

–End Quote 

The $11 billion is likely coming from other SKUs as we pointed out in our pre-earnings analysis “There is certainly a scenario where Nvidia’s GPUs are in such high demand that other SKUs can help make up for a delay on the much-larger GB200 NVL systems.” The additional evidence of this is the QoQ decline in networking as these systems would certainly drive QoQ growth in networking, making the clear readthrough that other SKUs are driving the $11 billion. 

When asked if there were bottle necks to consider with NVL 72 systems specifically, the CEO avoided discussing NVL72 systems and redirected to a more general discussion on the Grace Blackwell architecture. He also pointed toward single use cases that have “come online” rather than a “shipping in volume” discussion: “You've probably seen on the web, a fair number of celebrations about Grace Blackwell systems coming online and we have them, of course. We have a fairly large installation of Grace Blackwell goes for our own engineering and our own design teams and software teams. CoreWeave has now been quite public about the successful bring up of theirs. Microsoft has, of course, OpenAI has, and you're starting to see many come online. And so I think the answer to your question is nothing is easy about what we're doing, but we're doing great, and all of our partners are doing great.” 

TAKEAWAY: The NVL 72 systems are probably not shipping in volume right now, but should be soon (if they’re in production). This will run close to Blackwell Ultra’s launch. Demand will not be an issue rather it’s unusual to see two generations ship this close to one another, and it will require a smooth supply chain  

Custom Silicon Not a Threat for 4 Reasons 

Concerns around timing of shipping for Blackwell SKUs will be transitory yet concerns with custom silicon are here to stay. Since the very first day of covering the AI story in 2018, I have been fielding concerns around custom silicon as Google’s TPUs were all the rage in 2018. It was actually expected at the time that Google’s TPUs would transition to merchant sales and directly compete with Nvidia. Nearly seven years later, that has not materialized. 

Below, is how Jensen Huang describes the challenges around custom silicon. He points toward custom silicon being application specific whereas GPUs are more flexible (this part is key), the software layer, as well as the fact that a lot of custom silicon does not make it to full production. Because investors will be fielding this concern into the foreseeable future, I’ve noted the response and four reasons below: 

1. General versus application specific: “NVIDIA's architecture is general whether you're — you've optimized for unaggressive models or diffusion-based models or vision-based models or multimodal models or text models. We're great in all of it […] And so by definition, we're much, much more general than narrow” 
2. Nvidia is Everywhere: “So we're general, we're end-to-end, and we're everywhere. And because we're not in just one cloud, we're in every cloud, we could be on-prem. We could be in a robot. Our architecture is much more accessible and a great target initial target for anybody who's starting up a new company. And so we're everywhere.” 
3. Performance: “And if our performance per watt is anywhere from 2x to 4x to 8x, which is not unusual, it translates directly to revenues. And so if you have a 100-megawatt data center, if the performance or the throughput in that 100-megawatt or the gigawatt data center is 4 times or 8 times higher, your revenues for that gigawatt data center is 8 times higher.” 
4. What Nvidia Does is Hard – the Hardware and the Software is hard to replicate: “And the ecosystem that sits on top of our architecture is 10x more complex today than it was 2 years ago. And that's fairly obvious because the amount of software that the world is building on top of architecture is growing exponentially and AI is advancing very quickly. So bringing that whole ecosystem on top of multiple chips is hard.”

China Increases in Geo Mix for Q1 

An analyst pointed out that China will be increasing in geographic mix for Q1.  

Here is the concern from the analyst: “And I think there is a concern about whether U.S. can pick up the slack if there's regulations towards other geographies. And I was just wondering, as we go throughout the year, if this kind of surge in the U.S. continues and it's going to be — whether that's okay. And if that underlies your growth rate, how can you keep growing so fast with this mix shift towards the U.S.? Your guidance looks like China is probably up sequentially. So just wondering if you could go through that dynamic and maybe collect can weigh in.” 

This may be a moot issue, as management stated: “China is approximately the same percentage as Q4 and as previous quarters. It's about half of what it was before the export control. But it's approximately the same in percentage” but noting the concern here that this analyst thinks USA may not be able to fully make up for a China loss if the trade war heats up. I think if the NVL systems are shipping in volume, it will not matter as China is primarily helping to make up for the lag in these systems right now for the next 1-2 quarters.  

Conclusion: 

You might be asking yourself — why is the I/O Fund so focused on something the rest of the market does not seem to care about? There are two reasons … $100 billion pointed at one SKU is unheard of. Consider the iPhone took a decade to get to that revenue. We will not lazily assume it’s coming until I get “all systems are a go” signs across the board. You can’t put $100 billion into production and ship in volume without a splash in the large supply chain that builds these AI systems. There was no splash (yet). 

Secondly, we need Nvidia to break out of the range bound price it’s been in for 9 months (between $110 and $150). While many might be saying “whew, Nvidia beat/raised,” my stance is “where’s the catalyst that’s going to take this thing higher?” I had an answer for you a few months back – it was the NVL systems, which I stated would cause “fireworks.” To be range bound trading between $110 and $150 since May of 2024 is not exactly fireworks, and thus we need to look closely at whether Nvidia’s next move is to break out or break down. As stated, by being range bound, the valuation has become more attractive, and we are watching this piece closely. This is my number one reason for seeing a breakout above $150 right now (valuation). 

While many investors will rightly hold Nvidia through ups/downs (you’ll see IOF hold for years to come), our firm is competing with the world’s best tech portfolios – which means complacency is not something you’ll see here. Our Members greatly benefit from our need to compete as we question everything – especially if we are overweight any specific outcome.  

Please reference the most up to date trading plan here.

Please note: The I/O Fund conducts research and draws conclusions for the company’s portfolio. We then share that information with our readers and offer real-time trade notifications. This is not a guarantee of a stock’s performance and it is not financial advice. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis. Beth Kindig and the I/O Fund own shares in NVDA at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• Nvidia Q4 Financials: Beat Likely for Q4; Questions Persist for Q1
• Coinbase Posts Significant Growth Across the Board in Blowout Q4
• AI Hardware Suppliers Forecast Muted H1, Strong H2 — and What it Might Mean for Nvidia
• AMD Q4 Earnings: Q1 Decline in Data Center, Mgmt Drops AI Guidance

We’ve heard from a handful of management teams that are direct suppliers to Blackwell over the last four business days, and the tone has shifted to more of a H2 discussion for rack-level systems. There are a few possibilities for this shift in tone, such as these specific suppliers are falling out of favor with Nvidia, but given that a variety of suppliers are echoing one another, it’s more likely that Nvidia’s GB200s are not completely on time to ramp in volume for Q1 and at least for part of Q2 as “mid-summer” and “H2” has come up repeatedly. 

This is not a prediction, as reading tea leaves or a crystal ball is not the service we offer. What we offer is analysis – a way of connecting the dots to make more informed decisions. Right now, according to our analysis and earnings calls, the risk to the downside has increased that Nvidia is not shipping on time for the more complex GPU systems. Should the analysis play out, it simply means we will get Nvidia’s stock lower for the next leg higher.

The comments offered below are subtleties, as the last thing a supplier wants to do is get in the crosshairs of Nvidia. Yet, we’ve had a handful of suppliers’ report, and I’m struggling to find the expected strong commentary for the March quarter and there is also an absence of strong commentary for the June quarter (as it currently stands).

I’ve been on back-to-back earnings calls and offer the following excerpts in terms of this conclusion.

Astera Labs: Aries PCIe Gen 6 Tone Changed

It makes sense to start with Astera because they were the most direct in terms of discussing merchant GPUs (i.e., Nvidia) not being the driver for Q1.

Astera stated they are shipping pre-production quantities for its Scorpio P Series and Aries PCIe Gen 6 solutions to maximize GPU throughput. The company stated: “These programs are driving higher dollar content opportunities for Astera Labs on a full rack and full accelerator basis and we expect volume deployments to begin in the second half of this year.”

What this comment implies is both the Scorpio P Series and PCIe Gen 6 will deploy in H2, yet originally, PCIe 6.0 was expected to ramp with support initially offered in the GB200s. Back in March, Astera demo’ed PCIe 6.0 for a wide range of Blackwell products.

There was also indication back in the August call that Gen 6 was confirmed to be used in Blackwell’s GB200, and there were initial shipments: “We have started shipping initial quantities of preproduction orders of our PCIe Gen 6 solution, Aries 6. We ship and support our hyperscaler customers initial program developments that are based on Nvidia's Blackwell platform, including GB200.”

There was further confirmation in the November call that preproduction volumes were shipping: “At the recent 2024 OCP Global Summit, we demonstrated the industry's first PCIe Gen 6 fabric switch, which is currently shipping in preproduction volumes for AI platforms.”

As Nvidia investors, we keep a close eye on these suppliers as they often can flag any issues with unparalleled accuracy. Given we are nose-down on many earnings reports over the past year regarding Nvidia’s AI systems specifically, there’s a notable change in tone for Astera to go from preproduction volumes to now shipping in volume in H2.

The following was discussed in the Q&A section:

“Now, if you look into 2025, we see both contributing growth. The first half of the year will be more predominantly the AI – internal AI accelerator programs. But if you get into the back half of the year, the transition on the merchant GPUs will also be very strong for us. This is where you'll see the custom rack configurations start to deploy and that’s where we see a big dollar increase in our contemporary GPU with Scorpio starting to ramp.”

On one hand this is a positive, but on the other hand, it raises questions as to why merchant GPUs (Nvidia) will be stronger in the back half of the year as this does not match the anticipated timing originally described by Nvidia’s management team with the CEO stating, “Blackwell production is in full steam. I think we're in great shape with respect to the Blackwell ramp at this point.”

There could be an argument made it’s unique to Astera Labs, yet we have seen something similar echoed across a few management teams recently.

Thomas O'Malley   Barclays Bank

Super helpful. And then my follow-up was just — I think it was Mike's commentary on one of the first questions here on the call. You talked about kind of the year 2025 on the Aries side, talking about how in the first half of the year, you would see more internal AI efforts followed by the second half of the year being more merchant GPU. That comment was a bit surprising to me given we're going through a big product transition now at the large customer of yours. So is there any change in the way that you see the ramp of 2025 versus where you did before? I would have anticipated maybe the merchant GPU being a bit stronger earlier in the year. Just any reason behind those comments that caught me a little off guard.

Michael Tate   CFO

Sure. Yes, so the — first of all, the merchant GPU drives both Scorpio and Aries. So the big incremental piece of the merchant GPUs is the score field content which is all new for us. The designs that we have are complex in nature, they're all new. So the — to get them — to productize and ramp we're looking at that to start off in the back half of the year. Right now, in the first half of the year's preproduction. These are all for custom configuration. So the customization adds a little bit of lead time to the volume rates.

Coherent: Tone on Pluggable Optics Versus Active Cable

In our analysis published last week, it was stated: “As we look at larger AI systems ramping this year, there is some evidence that issues around single-rack Blackwell systems have been resolved yet multi-rack interconnects continue to see issues with overheating. Even though Nvidia has stated they are on schedule with Blackwell, the B300s and next generation Rubin will very likely require more fine-tuning.”

However, there was a sudden change in Coherent’s tone around fiber optics versus copper, which we outlined here:

“Looking beyond traditional pluggable optics, there is an increasing amount of discussion around co-packaged optics (CPOs), which places the optical transceivers directly on the chip package, rather than using separate optical modules. This results in faster data transmission, reduced latency and higher bandwidth. This may be the best of both worlds: the performance of optical yet with reduced power consumption. Tracking this is especially important as since we last covered copper/Semtech, there have been reports that copper is “causing concurrent issues with overheating and glitching” with rumors Nvidia will launch a CPO switch at the upcoming GTC. That could mean Coherent will be a lead supplier for the anticipated CPO switch – we will be monitoring this closely.”

This later led to Semtech stating they expect sales of CopperEdge products to be lower than the $50 million guidance provided in the earnings call four weeks ago. Our team’s spidey senses were up following Coherent’s report (per the paragraph above), and thus, we respected the stop we provided to our Members in weekly webinars. The stop was hit on Friday with a trade alert stating the story had changed about 30 mins before market close. The context is important as it communicates that Semtech was also caught off guard as to changes with the server rack configuration with a sudden, unforeseen change in the last four weeks.

Regarding Coherent, if they were the beneficiary of pluggable optics replacing copper, one would assume a strong Q1. Yet when asked about the flat QoQ revenue on the call, the CEO stated it was not due to the AI-related segments: “It's pretty straightforward. We expect datacom and telecom to be up sequentially. And we expect the rest of our industrial related businesses to be down sequentially. And that net at the midpoint to be flat.”

Perhaps AI-related segments are remaining strong enough to not miss guidance, yet why are they not able to offset industrial weakness for sequential growth. That piece was left unanswered.

Read-through for Credo?

One potential read-through is that Credo is in a better position to gain the business following Semtech’s announcement with its AECs for AI backend networks. The company also offers optical DSPs and PAM4 solutions. The possibility of Credo gaining this business is an outcome we have been prepping for by owning a wide array of suppliers. Read more here.

With that said, it’s the sudden change of information from four weeks ago – announced Friday after market close — that marks a notable shift our Members should be aware of.

Power Management Integrated Circuits (PMICs):

This component is where the rubber meets the road as the thermal management issues Blackwell has been rumored to face are addressed with PMICs.

Monolithic Power Systems reported on Friday and had many comments about their ramp being H2 weighted: “Yes. Just to add a little bit of color to how we see the year rolling out, we believe that, we will be off to a slower start in the first half of the year. But as the year develops, the customer base is expected to broaden as hyperscalers launch their new products. We have multiple product ramps with both existing customers as well as with these new hyperscalers. So as Michael just said, we believe, it's likely to be a flattish year, but we believe that from a quality and supply availability perspective, we're in very good shape.”

On one hand, some analysts do not think MPS is designed into the Blackwell systems (Keybanc) whereas Truist is “highly confident” MPS is designed into the systems. According to this report, I lean with Keybanc that MPS is designed into Blackwell Ultra.

“(11/18) KeyBanc lowered the firm's price target on Monolithic Power to $700 from $1,075 and keeps an Overweight rating on the shares. The firm believes Monolithic Power will lose significant market share on Blackwell with the ramp of GB200/B200, as Hopper PMIC overheating issues have persisted on Blackwell. KeyBanc expects IFX to be the primary supplier on Blackwell, with Renesas having secondary share. While Monolithic Power is attempting to requalify, the earliest this likely could happen would be with Blackwell Ultra in the second half of 2025.”

“(12/16) Truist analyst William Stein lowered the firm's price target on Monolithic Power to $762 from $887 and keeps a Buy rating on the shares. The firm reiterated a cautious semiconductor and artificial intelligence sector view, but is more constructive on Nvidia (NVDA) and Monolithic Power (MPWR) while more cautious on Tesla (TSLA). While Monolithic Power's print position at Nvidia is in doubt today, Truist is "highly confident" that Monolithic is designed in to Blackwell, the analyst tells investors in a research note. The firm cut enterprise data revenue estimates for Q1 significantly, but highlights Monolithic's "likely top position across a breadth of AI customers, supporting solid growth over the next few years."

Yet another PMIC supplier that we covered on the Advanced side on Feb 5th, is stating that “We know that we're not the only one players going after this, so the share is still to be determined. The final design, everything is still to be finalized as well, too. So we're working closely with the customer. Again, at this point, we can say that we're targeting for a mid-year launch.”

This company is either discussing Blackwell Ultra with the B300s, or they are implying GB200s are delayed. After hearing from more component suppliers, the likelihood they are actually discussing Blackwell (and not Blackwell Ultra) has increased since we wrote the analysis last week.

Conclusion:

Investors should have a plan for this – ride out any weakness with the understanding there is $300 billion in capex and growing pointed right at AI hardware, or have a more active stance which includes risk controls such as stops on positions, hedging, etcetera. For the IOF, we hedge while having predefined price targets and predefined stops. For the most part, should the analysis we’ve presented above play out, then our goal will be to buy quality suppliers at lower prices. I firmly believe the suppliers that move Blackwell forward with be greatly rewarded, and the hard work we are putting in will be rewarded.

Direct liquid cooling suppliers report today at the bell (SMCI) and tomorrow before market open (Vertiv). Let’s see if they are aligned (or not) with the other suppliers offering a mixed Q1 outlook.

Regarding Nvidia, we recently trimmed 6% to re-allocate to Nvidia AI hardware suppliers. If we break $126 and/or $113, it’s my understanding we will trim more with the goal of loading back up sub-$100 and perhaps even sub-$90. We will discuss this more in our Advanced webinar this Thursday.

Please note: The I/O Fund conducts research and draws conclusions for the company’s portfolio. We then share that information with our readers and offer real-time trade notifications. This is not a guarantee of a stock’s performance and it is not financial advice. Please consult your personal financial advisor before buying any stock in the companies mentioned in this analysis. Beth Kindig and the I/O Fund own shares in NVDA at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• Astera Labs Q4 Earnings: Strong Commentary for H2 Ramp, What it Could Mean
• Coherent FQ2: Networking Segment up 56% YoY, Keep an Eye on Co-optics for H2
• AMD Q4 Earnings: Q1 Decline in Data Center, Mgmt Drops AI Guidance
• Amphenol: High-Performance Interconnects for the AI Ecosystem