Moo Live

was live ·

Broadcom Q2 2026 earnings conference call

Key Takeaways (AI-Generated)

Financial Performance

- Q2 2026 total revenue reached a record $22.2 billion, up 48% year-over-year, exceeding guidance

- Q2 operating margin hit a record 67%, and Adjusted EBITDA reached a record 69% of revenue

- AI semiconductor revenue hit a record $10.8 billion, up 143% year-over-year, driving growth

- Q2 free cash flow reached a record $10.3 billion, representing 46% of revenue

Business Highlights

- Announced a long-term agreement with Google to develop and supply multiple generations of TPUs

- Partnership with Meta to deliver multiple generations of MTI XPUs with a 3-gigawatt deployment

- Creating an AI XPU platform with Apollo and Blackstone to deploy over 20 gigawatts of capacity

- AI semiconductor bookings exceeded $30 billion against $10.8 billion shipped in Q2

Financial Guidance

- Q3 2026 consolidated revenue expected to be $29.4 billion, up 84% year-over-year

- Q3 AI semiconductor revenue expected to accelerate to $16 billion, up over 200%

- Full year 2026 AI semiconductor revenue expected to be $56 billion, up 180%

- Fiscal 2027 AI semiconductor revenue guidance reiterated to exceed $100 billion

Opportunities

- AI semiconductor revenue expected to double from first half to second half 2026

- Strategic partnerships with major customers for multi-gigawatt deployments through 2028

- Leading technology in XPUs, networking solutions and co-packaged optics development

- AI XPU platform with Apollo and Blackstone valued at $35 billion first tranche

Full Transcript (AI-Generated)

Operator

We announced Kirsten will be retiring June 12th and today we have joining us our incoming Chief Financial Officer, Amy Teener. Thank you, Kirsten, for your leadership over the past 12 years. Broadcom distributed a press release and financial tables after the market closed describing our financial performance for the second quarter of fiscal year 2026. If you did not receive a copy, you may obtain the information from the Investors section of Broadcom's website at broadcom.com.

This conference call is being webcast live and then audio replay of the call can be accessed for one year through the Investors section of Broadcom's website. During the prepared comments, Hawk and Kirsten will be providing details of our second quarter fiscal year 2026 results, guidance for our third quarter fiscal year 2026, as well as commentary regarding the business environment. We'll take questions after the end of our prepared comments.

Please refer to our press release today and our recent filings with the SEC for information on risk factors that could cause our actual results to differ materially from the forward-looking statements made on this call. In addition to US GAAP reporting, Broadcom reports certain financial measures on a non GAAP basis. A reconciliation between GAAP and non GAAP measures to the extent possible is included in the tables attached to today's press release. Comments made during today's call will primarily refer to our non GAAP financial results. I will now turn the call over to Hawk.

Hawk

Thank you, Jay. Thank you everyone for joining today. In our fiscal Q2 2026, total revenue reached a record $22.2 billion, up 48% year on year, above our guidance on strength in the AI semiconductors. Q2 operating margin was a record 67% and Adjusted EBITDA was a record 69% of revenue, which was above our guidance. Even as our revenue scales are massively driven by AI, our operating and EBITDA margins remain strong and stable.

Turning to semiconductors, Q2 revenue was a record $15 billion as we grew 79% year on year. Driving this growth was AI Semiconductor revenue at a record $10.8 billion, up 143% year on year and above our outlook. Networking represented almost 40% of our Q2 AI revenue demand for XPUs and networking is simply insatiable.

During the quarter bookings for AI semiconductors were over $30 billion against the $10.8 billion we shipped in the second-half of 2026. We expect AI semiconductor revenue to double from the first half we shipped this year. Consistent with this trend, in Q3, we expect AI semiconductor revenue to accelerate to $16 billion, up over 200% year on year.

For the full year 2026, we expect to achieve AI semiconductor revenue of $56 billion, up approximately 180% from fiscal 2025. Now we expect this momentum to continue into fiscal year 2027 and reiterate our AI semiconductor revenue guidance to be in excess of $100 billion. We expect AI semiconductor revenue growth to continue in fiscal 2028 based on the following initiatives we have with our 6 core customers.

As you are aware, we announced in April that we entered into a long-term agreement to develop and supply multiple generations of TPUs and AI networking. Our relationship continues to be strategic and very substantial as we continue to deliver vastly superior technology and execution compared to other alternatives. This ability to provide differentiated value to Google ensures that our business will sustain and grow for the foreseeable future.

For Entropic, as you know for 2026, we are providing access to Broadcom TPU-based compute of over 1 gigawatt. In April, we entered into an agreement to enable Entropic to access another 5 gigawatts of next-generation TPU-based compute beginning in 2027. For OpenAI, we have delivered silicon and we are on track for production late 2026. We have a contractual commitment to deploy 1.3 gigawatts in 2027 as part of the larger 10 gigawatts by 2029 agreement we announced last year.

For Meta, in April, we announced the partnership to deliver multiple generations of MTI XPUs and under this agreement we expect to deploy 3 gigawatts through the end of 2028. The initial order for one gigawatt, which includes XPUs and our networking, has been received and we'll start delivery in the second half of 2027. For our other two customers, we expect shipments to begin late 2026 and accelerate into 2027. To date, we have received purchase orders totaling $6 billion.

While we have significant IP and execution leadership in XPUs, networking is key to building scalable XPU and GPU clusters and here in networking we hold at least one generation of technology and product leadership. For scale-up within racks, we enable direct-attached copper based on industry-leading 200G and 400G standards, driving co-packaged copper with Ethernet and PCIe switches.

For scale-out between the racks, we have been shipping the industry's only 100-terabit Ethernet switch, the Tomahawk 6, for over a year. We will now be taping out our next-generation 200-terabit switch this quarter. And in CPO—co-packaged optics with 1.6-terabit DSPs, CW and EML lasers—we are the de facto standard in the industry to extend AI clusters across data centers.

We remain the industry leader with our Jericho 3 and Jericho 4 fabric solutions and are enabling the world's largest deployments at multiple hyperscalers. Our strategic vision is to bring together Broadcom's leading technology and investor partners with the strongest balance sheets to deliver at scale sufficient compute capacity at the lowest cost and power for the leading AI frontier labs, including Entropic and OpenAI.

To deliver this vision, we are creating the AI XPU platform with Apollo and Blackstone and other leading investors to deploy more than 20 gigawatts of compute capacity through 2028. The first tranche of this platform, valued at $35 billion, is currently being launched by Apollo.

Now turning to non-AI semiconductors. Q2 revenue of $4.2 billion was up 6% year-over-year. Bookings during the same period exceeded $6 billion, which is a clear indication we are on the path towards a full cyclical recovery. Broadband, server storage, and enterprise networking together were up, partially offset by a seasonal decline in wireless. Consistent with this trend, in Q3 we forecast non-AI semiconductor revenue to be approximately $4.5 billion, up 12% from a year ago.

In summary, we expect Q3 semiconductor revenue to be $20.5 billion, up 124% year-over-year. Let me turn to the Infrastructure Software segment. Q2 software revenue of $7.2 billion was up 9% year-over-year, in line with our guidance. Bookings continue to be strong as we sustain ARR growth of 17% year-over-year. For Q3, we forecast software revenue to be approximately $8.9 billion, up 31% year-over-year.

We just released VMware Cloud Foundation 9.1, focusing on improving infrastructure efficiency, security, and support for enterprise AI inferencing workloads amid strong global server demand. The deployment of VCF 9.1 for on-premises cloud computing is extremely robust, driving strong revenue growth. This release adds heterogeneous compute support across GPUs and CPU architectures, including AMD, Intel, and NVIDIA platforms, enabling enterprise cloud customers to run AI, Kubernetes, and traditional virtualized workloads on a common private cloud environment.

So to sum it up, for Q3 2026, we expect our consolidated revenue to grow to $29.4 billion, up 84% year over year. We expect operating margin to be stable at approximately 67% of revenue and Adjusted EBITDA to be at approximately 68% of revenue. And with that, let me turn the call over to Kirsten.

Kirsten

Thank you, Hawk. Let me now provide additional detail on our Q2 financial performance. Consolidated revenue was a record $22.2 billion for the quarter, up 48% from a year ago. Gross margin was 77.1% of revenue in the quarter, down 230 basis points year over year as semiconductors became a larger proportion of our product mix.

Consolidated operating expenses were $2.2 billion, of which $1.6 billion was R&D. Q2 operating income was a record $14.9 billion, up 52% from a year ago. Note that even with the decline in gross margin, operating margin increased by 200 basis points year over year to 67.3%, as operating expenses remained relatively flat. Adjusted EBITDA of $15.2 billion, or 69% of revenue, was above our guidance of 68%.

Now a review of the P&L for our two segments, starting with semiconductors. Revenue for our Semiconductor Solutions segment was a record $15 billion, with growth accelerating to 79% year over year, driven by AI. Semiconductor revenue represented 68% of total revenue in the quarter, and AI semiconductor revenue represented 49% of total revenue. Gross margin for our Semiconductor Solutions segment was approximately 70%.

Operating expenses of $1.2 billion reflected increased investment in R&D for leading-edge AI semiconductors and represented 8% of revenue. Semiconductor operating margin of 62% was up 460 basis points year over year, reflecting our strong operating leverage. Now moving on to infrastructure software. Revenue for infrastructure software of $7.2 billion was up 9% year over year and represented 32% of total revenue.

Gross margin for infrastructure software was 93% in the quarter, and operating expenses were $1 billion in the quarter. Q2 software operating margin was up 310 basis points year over year to approximately 79%. Moving on to cash flow: free cash flow in the quarter was a record $10.3 billion and represented 46% of revenue. We spent $231 million on capital expenditures.

We ended the second quarter with $19.6 billion of cash, compared to $14.2 billion in the prior quarter. We ended the second quarter with inventory of $4.3 billion as we continue to secure supply to support strong AI demand. Our days of inventory on hand were 86 days in Q2, compared to 68 days in Q1, in anticipation of accelerating AI semiconductor growth in the second half of the year.

Turning to capital allocation: In Q2, we paid stockholders $3.1 billion in cash dividends, based on a quarterly common stock cash dividend of $0.65 per share. Now moving to guidance: our guidance for Q3 is for consolidated revenue of $29.4 billion, up 84% year over year. We forecast semiconductor revenue of approximately $20.5 billion, up 124% year over year.

Within this, we expect Q3 AI semiconductor revenue of $16 billion, up over 200% year on year. We expect Q3 infrastructure software revenue of approximately $8.9 billion, up 31% year on year. Moving on to margins, as the proportion of AI revenue significantly grows in Q3, we expect Q3 consolidated gross margin to be down to approximately 74%.

This decline in gross margin does not represent a structural change in semiconductor margin, rather it reflects product mix between semiconductors and infrastructure software. Regardless of the impact to gross margin, we expect Q3 operating margin to be 67%, which is flat quarter on quarter, demonstrating our strong operating leverage. We highly recommend that investors model semiconductor and infrastructure software margins separately to properly reflect the impact of changes in total revenue mix going forward.

We expect the non-GAAP tax rate for Q3 and fiscal year 2026 to be approximately 16% due to the impact of the global minimum tax and the geographic mix of income compared to that of fiscal year 25. In Q3, we expect the non-GAAP diluted share count to be approximately 4.94 billion shares excluding the impact of potential share repurchases. That concludes my prepared remarks. Operator, please open up the call for questions.

Operator

Thank you. To ask a question, you will need to press *11 on your telephone. To withdraw your question, press *11. Again, due to time restraints, we ask that you please limit yourself to one question. Please stand by while we compile the Q&A roster. And our first question will come from the line of Harlan Sur with JP Morgan. Your line is open.

Harlan Sur

Yeah, good afternoon. Thank you for taking my question. Thanks for all your support. Kirsten and Amy, welcome to the team. First, just the housekeeping, quick housekeeping item—Hawk, on this fiscal year AI sort of 2X growth second-half over first half, that would put AI revenues over $60 billion with sequential growth in Q4. But you gave us this $56 billion number, which is only like 1.5X, you know, half-over-half growth, with Q4 AI actually being down sequentially. So if you could just help us kind of square the numbers there.

And then for my real question—you know, back in December of last year, you talked about an $73 billion AI backlog over the next 18 months. If you spread that number linearly over six quarters, but we know that the backlog is always more front-loaded over the first four quarters, right? And sure enough, you're going to deliver around 80% or more of that backlog in this fiscal year, or the first four quarters, just given the strength of all your programs, the broadening of the customer base, accelerating year-over-year trends in your AI shipments, and all the multi-gigawatt partnerships you just articulated today—most of which are set to ramp up next year. Is it fair to assume that your 18-month AI backlog, from the second half of this year through all of fiscal 2027, sits at $200 billion or better?

Hawk

That's a very complicated set of number questions to begin with. Let's start with fiscal 2026—doing the math, basically 2X the first half. We shipped about $19 billion in total AI revenue in the first half—you can be precise on that. So if you take what I indicated and double that in the second half, you get pretty much in the range of what we're talking about, which is around $56 billion, Harlan. So those numbers still tie up very, very well.

Now you'll pick a question on the second now which you're going to need a very detailed analysis office. Yeah, we keep the momentum going as we expect to see in 2027. What we will see in 2027 is continued growth of the level we're talking about. And if you drive on that basis of what we're seeing here, it’s almost 2X what 2000 was, in the range of 2X what 2026 will be. I think you will easily see that 2027 will exceed very easily $100 billion in 2027, which is pretty much what we indicated last quarter.

And we are continuing to say that it will be over $100 billion in 2027. So in that sense, if anything else, it might be based on what we're doing—very much on track, if not stronger—but we're not trying to guide you every quarter on what 2027 would be like. So we basically say it continues to be in excess of $100 billion in 2027, but it is on the same trajectory as we are seeing in the back half of 2026.

Harlan Sur

Got it. OK. Thank you, Hawk.

Operator

One moment for our next question, and that will come from the line of Blaine Curtis with Jefferies. Your line is open.

Blaine Curtis

Hey, good afternoon. Thanks for taking my question. Hawk, I wanted to ask you—intra-quarter, you had that 8-K with the long-term agreement with Google. I think obviously you're probably not going to tell me what the total value is there, but I think there's a lot of concern about share within that customer. I was just kind of curious—now that you have this agreement, maybe you could speak a little bit more to your confidence and whether there’s upside with that customer. Is it a fixed amount, or is there share? Is there any way you can add some color to that agreement that came out?

Hawk

Well, you know, it's a very, very strong agreement, and it basically reflects the strength of the partnership we have—simply because of the products we do, the multi-gen products, and the intellectual property we deploy into this whole program. To answer your question specifically, it's a commitment that is very substantial in dollars—a very, very substantial amount of dollars.

Now we also accept the fact that while we like to win every design in that program, we also accept the fact that given the rate, the growth of consumption of and development and consumption of AI compute even by our partner Google, that we fully expect that there will be some diversity of sources for them. But our commitment from them is a very substantial dollar amount.

Blaine Curtis

Thank you

Operator

One moment for our next question and that will come from the line of Ross Seymour with Deutsche Bank. Your line is open.

Ross Seymour

Hi. Thanks for the asked question and Congrats to both Kirsten and Amy. Kirsten, on the gross margin side of things, I know you talked about it going down due to the mixed dynamics within the semis versus the software side. But given the strength and the software side in the quarter, it seems like the gross margins falling a little bit harder. So behind the scenes, can you just talk a little bit about what the drivers within semis are? Is that the XPU versus the networking side of things? And is that trend likely to continue next year? Are there rack scale versus chip scale, all those sorts of dynamics, Any color you could give on that would be helpful.

Kirsten

Yes, certainly as our semiconductor business grows, just to reiterate on a consolidated basis relative to our software business, you're going to have a decline in margins, right, you'll have compression. But remember that we're it's accretive because we have strong operating leverage, right. So our operating margins will stand up a bit over time to that. Within semiconductors, we've always said our ASICS and TPUs, some of the wireless business has lower margins. So as the TPUs continue to accelerate, there'll be pressure overall on margins. But the connectivity side, the AI networking side of the business has very rich margins. So it'll offset it somewhat as we go.

Hawk

I mean, Ross, as Kirsten said in her remarks, structurally the semiconductor margins remain very stable and very solid. It's the mix—the particular mix between software and non-AI versus the very, very high, rapidly growing AI semiconductor segment—that is just diluting gross margin on the rack side compared to the chip side of things. Is that all clarified now?

Ross Seymour

No, no, no—racks is only chip business only. We only do chips, only chips.

Hawk

Perfect. Thank you.

Operator

One moment for our next question, which will come from the line of Ben Reitzes with Melius. Your line is open.

Ben Reitzes

Yeah, hey guys, thanks. Appreciate it. Wanted to ask about 2027 Hawk with regard to—you know, previously we've talked about the TAM being capped. Well, actually, it’s kind of a longer-term question. You’ve talked about the TAM being $10 billion to $20 billion and so on. It seems that one of your competitors recently mentioned that the TAM per gigawatt is going up significantly throughout the decade, and it appears it wasn’t just due to infrastructure—it was also driven by compute and networking components and other factors.

Perhaps you’re familiar with that comment Jensen made, where overall infrastructure spending is moving from around $50-something billion toward $100 billion, and the compute content is increasing substantially. Are you seeing the same trend over the long term, potentially acting as an accelerator beyond what you’ve already outlined regarding your TAM per gigawatt? How are you thinking about that? Thanks a lot.

Hawk

Sure. Well, I think the accelerating part—if you talk about power—realize one thing: it’s the dollars per gigawatt, the content dollars of Broadcom per gigawatt. It's not accelerating that much because you're creating chips where each individual chip is driving higher and higher power. So you're using fewer chips, although the average selling price (ASP) of each chip is increasing. Therefore, dollars per gigawatt—billions of dollars per gigawatt—is relatively stable, but the number of gigawatts will keep growing and will accelerate, as I think some of our remarks indicate.

And that's exactly what we’re saying: the amount of gigawatts required—measuring compute capacity by the number of gigawatts—is growing very fast. We're seeing this, particularly to the point where, for even two of our customers, Entropic and OpenAI, for whom we're building this platform to enable sufficient compute power, the capacity measured in gigawatts of power is far ahead of what we fully anticipated just six months ago.

And that’s just these guys. We don’t even discuss consumption beyond the XPU platform. We’ve announced here workloads from our other customers—Google’s internal workloads, Meta’s workloads, and other customers including the two mentioned earlier. Factor all that in, and you're talking about total gigawatts that, if you look at 2027 or 2028, will continue to grow. In fact, we expect 2028 to show substantial growth compared to our current 2027 forecast.

Ben Reitzes

Thank you.

Operator

One moment for our next question, which will come from the line of Timothy Arcuri with UBS. Your line is open.

Timothy Arcuri

Thanks a lot. I wanted to ask you about supply and your ability to secure incremental wafer and HBM volumes. When I look at some of your competitors, they seem able to pull, say, $20 billion out of thin air and secure additional wafer supply. So I’m wondering: do you feel confident that if a customer approaches you, you can secure upside capacity in terms of wafers and HBM? And are you starting to consider using other foundries to add more optionality to your supply chain? Thanks a lot.

Hawk

Getting supply is not just about dropping money though, that does help. No, we are very comfortable that we have been able to secure supply of the types you mentioned for our needs—2627 is working on 28 and 29 right now, right? But if a customer comes to you and wants incremental supply, are you able to go to your suppliers and get it the way that it seems like some of your competitors are?

Timothy Arcuri

All customers have been coming to us incrementally over the last few months. Expect that to continue and, by and large, yes.

Hawk

OK, thank you.

Operator

One moment for our next question, and that will come from the line of Stacy Rasgon with Bernstein Research. Your line is open.

Stacy Rasgon

Hi, guys, thanks for taking my question. Hawk, you gave some gigawatt shipment targets for next year for your various customers. I just want to know—are those any different? Do they fully reflect or contemplate any changes from what you said last quarter? I think you'd said that was like close to 10 gigawatts you'd be shipping in '27. And can you just help us shape the year? It sounded to me like you expected that to be more back-half loaded in '27 given the shape of the ramps, but most importantly, is there any change—is it more gigawatts, less gigawatts, or the same gigawatts versus what you were suggesting last quarter?

Hawk

Well, fortunately, good question. Yeah, for 2027 we indicated about 10 gigawatts of shipments in ’27. That target remains very much intact. They will be shipping 10, and we’re planning to ship 10 gigawatts in ’27—nothing has changed to that extent, yes. And this really sets up an interesting trajectory into 2028 with this second-half momentum. So for 2028, we expect significantly more gigawatts.

Stacy Rasgon

Got it, that’s helpful. Thank you.

Operator

One moment for our next question, which will come from the line of Jim Schneider with Goldman Sachs. Your line is open.

Jim Schneider

Good afternoon. Thanks for taking my question. I was wondering if you could comment a bit on the profile of your networking business as we move through fiscal years 2026 and 2027. About 40% of AI revenue came from networking this quarter. Would you expect that percentage to decline somewhat as some of these custom ramps ramp up toward year-end and into early next year? Or would you expect to remain at the upper end of that range? Also, could you discuss when you anticipate optical CPO revenue becoming meaningful? Thank you.

Hawk

That’s a hell of a great question, Jim. It’s just very difficult to answer because of how unpredictable it is. There are quite a few moving parts here—one of which is this: as more and more of our customers shift toward XPUs, XPUs naturally use far more of our networking components across the board. That’s obviously great for us and drives higher consumption. However, it also means we’ve been able to sell networking solutions to non-XPU customers as well, and that portion will dilute the overall growth rate.

And what I mean by this 40%—I consider it to be very much at the upper end of what’s reasonable. It's almost a scenario where we're shipping a lot of networking gear to non-XPU customers. While the growth of XPUs is clearly enabling us to expand this networking business tied to our XPUs, pushing us toward that 40% mark, I see that as probably the highest that percentage of total AI revenue could go—not for the first time. I’ve previously indicated that a more typical or expected share of networking within total AI revenue would be closer to around 30%.

Operator

One moment for our next question. That will come from the line of Tom O'Malley with Barclays. Your line is open.

Tom O'Malley

Hey, thanks for taking the question. So I noticed with the most recent deal with Entropic that you guys are using Broadcom chips as a backstop for the deal. Do you expect more deals like this in the future? And as you start to see the AI environment evolve, are you thinking about financing these deals differently going forward—will you continue to use chips or offer anything else along those lines? Thank you.

Hawk

Can you repeat that question, especially the beginning part? I didn’t quite catch what you were saying there. I don’t want to answer incorrectly—sorry, Hawk.

Tom O'Malley

Essentially, the most recent deal with Entropic is being backstopped by Broadcom chips. Do you think you’ll see more deals structured this way in the future? And do you have any comments on how you might finance deals involving large AI models going forward? Thank you.

Hawk

I have to correct you on that. I will deal then I'll deal with Entropic and that we basically talk about and basically release this closing on 8K recently as the deal we did with Entropic is we use our TPU chips that we developed to provide the compute capacity to Entropic. We want that it wasn't backstop in that sense. We want the ones providing the chips to Entropic. We were the ones providing the compute capacity to Entropic.

Tom O'Malley

Thank you.

Operator

One moment for our next question that will come from the line of CJ Muse with Cantor Fitzgerald. Your line is open.

CJ Muse

Yeah, good afternoon. Thanks for taking the question. I guess Hawk, you know, in recent years you've talked about really focusing your efforts on very large XPU platforms. And I'm just curious—we're seeing many kinds of XPU-attached derivatives, you know, across interconnect, storage, and others—and I'm wondering if there are any niche programs there that are whetting your appetite.

Hawk

Well, no, I don't think so. I think our business model is actually very, very straightforward, which is we are developing XPUs—custom AI accelerators—for use by customers who are pretty much all LLM developers, whether for training or inference. We are also creating a portfolio of critical components to enable these XPUs—and even GPUs—to be clustered and deliver better performance, and that continues to be the model we follow. That is, we develop chip technology in the form of AI computing accelerators—which we call XPUs—as well as networking chips that cluster them together, including switches, PCIe connectors, DSPs, lasers, NICs, and routers.

And that’s very much still the model we employ in semiconductors. As you can see from our financial model and the direction of our programs, we remain focused on driving toward a chip-based business model through the technologies we provide. What we’re doing to enable some of these LLM players to access the volume of compute capacity—the large gigawatts of computing power—they need to scale up their models is, as I announced here today, creating—in partnership with firms that have the strongest balance sheets—a vehicle to essentially fund these chips for LLM players who might otherwise struggle to gain access to our technology, which offers them the lowest power consumption and lowest cost.

Operator

One moment for our next question. That will come from the line of Atif Malik with Citi. Your line is open.

Atif Malik

Hi, thank you for taking my question. I have a question on your infrastructure software business. Are you seeing any impact from AI, specifically agentic AI, on your software growth and renewals? And could you discuss the long-term growth outlook for that business?

Hawk

Well, we’re not seeing any negative impact. If anything, as I mentioned earlier, the high volume of CPU core counts being sold alongside GPUs is driving accelerated growth in our VMware business. As you saw in Q3, we’re experiencing accelerated growth, and we expect this trend to continue over the next several quarters as demand builds long term. Given the nature of our infrastructure software products—which sit very close to the hardware, essentially at the hypervisor layer where our products operate—we don’t anticipate any adverse impact on our software offerings.

Operator

And one moment for our next question, which will come from the line of Edward Snyder with Charter Equity Research. Your line is open.

Edward Snyder

Thanks a lot. This is very interesting because the gigawatts you’ve outlined for different customers clearly show that the two offering consumer-facing versions of AI—whether you call it CSP, CFC, or whatever—namely Entropic and OpenAI, have very large gigawatt commitments in the out-years. I understand part of that is catch-up since they started later, whereas your earliest customers have been doing this for quite some time. But even part of Google’s commitment involves offering cloud services to other parties as well.

So are we seeing a shift here? I know that initially a lot of the XPUs in the AI services went to the hyperscalers with their own customer workloads. We've talked about that ad nauseam. And now you're seeing AI finally hit the enterprises, and you've seen cloud take off with the programming that's sweeping everybody. Do we expect, then, that there's going to be this big second wave of demand driven as AI starts hitting enterprises—either through them gaining access to it or finding usable tools? Because the numbers you're citing here are significantly different for these two classes of customers.

Hawk

Well, there's a very interesting point here, and you may very well be right that enterprise adoption of AI is still relatively early in the game. But having said that, what we're also seeing is that a lot of what enterprises are consuming involves tokens—they're buying a lot of these tokens from platforms via product APIs or platform APIs. They’re pulling from the same platforms run by the major customers we’ve discussed before: Anthropic, OpenAI, Gemini, and others among the leading large model providers. And that’s where I think the vast majority of token consumption is tied to those LLMs—and as these LLM providers productize their frontier models, whether it’s Opus 4.7, ChatGPT 5.5, or Gemini Pro 1.5, it all circles back to the same core demand for compute capacity—the kind we provide to all these players.

And so even the growth in enterprise demand that we’re now starting to see—as enterprises begin consuming AI tokens for their own workloads and productivity use cases, much like consumer users do—is coming from the same sources: these few dominant platforms. That’s what’s driving what I’d call an insatiable surge in demand for compute capacity, which we’re experiencing firsthand. And we see this trend continuing not just through 2027 but also into 2028. This is shaping up to be a highly sustainable and steepening trajectory of demand.

Edward Snyder

So if I may—if this is indeed happening, doesn’t it change the dynamic we discussed earlier? You previously mentioned around seven key customers for your XPUs. But if enterprises are now accessing AI through cloud platforms—like Google offering TPU-based cloud services—that opens XPU access to countless smaller companies that don’t meet the threshold to develop their own ASICs or partner directly with Broadcom on custom silicon via these platforms. Why wouldn’t that be the case?

Hawk

Well, I suppose that’s possible, but the reality is this: most AI-generated compute capacity is delivered via a SaaS model. APIs are pulled from the cloud—whether from Bedrock, Vertex, Azure, or first-party offerings—and it’s still fundamentally provided through the cloud. So, at the end of the day, when it comes to compute capacity—the very thing we supply—it originates primarily from those few large frontier model developers and the products they create to serve both consumers and enterprises globally.

The source of demand stems from those frontier model labs developing the products that consumers, enterprises, and companies like ours ultimately consume. What we’re doing is providing compute capacity directly to that demand source—rather than going to a specific company or bank and trying to sell them XPUs, after which they’d need to build infrastructure, develop software stacks, and write applications to run AI themselves. I’m sure some enterprises are starting to do this now, but they’re few and far between—it’s still very early days for that approach.

The truth is, most of the demand is coming from the frontier model players who are building products—things like code assistants or engineering-specific vertical applications—which all originate from the same small group of companies developing frontier models. It’s not really coming from 100,000 companies directly trying to buy XPUs—or, for that matter, GPUs. It’s simply not happening at that scale yet.

Operator

One moment for our next question, and that will come from the line of Joe Moore with Morgan Stanley. Your line is open.

Joe Moore

Great, thank you. You mentioned $30 billion in AI bookings this quarter, which is a lot—presumably relative to shipments this quarter and next. Can you discuss the dynamics here? Why is there so much backlog now? Or, as you’ve indicated, you can flex supply upward—but why are bookings so high this quarter relative to revenue?

Hawk

Well, that’s driven by enormous compute demand. We’re seeing many large customers—particularly a handful of major ones—now realize that securing compute capacity requires lead time. They need to plan ahead thoughtfully. And it’s not just about ordering wafers to get chips or memory to ensure HBM or DRAM availability. They’re also saying, ‘I need power—I need PowerShell.’ So all of this involves forward planning. The bookings we’re receiving aren’t for immediate delivery. Some hope to receive sooner, but the reality they all accept is that they need to align quite a few other elements before deployment can happen.

But they’re placing orders early—they’re placing orders now—and they’re placing them at a scale that reflects very strong demand. This gives us significantly more visibility than we’d normally have in the semiconductor industry. Our current visibility extends all the way to 2028. Three months ago, I can tell you, visibility only reached into 2027; today, it runs through 2028. That’s part of—and a major reason why—we’re building this XPU platform: it’s truly the foundational platform designed to scale up and deploy the necessary capacity for our leading-edge model customers. As you’ve seen in some of their financial disclosures and real-world experiences they’ve shared with you, this is driving massive token consumption from the compute capacity we’re providing them.

We now benefit from substantial lead time, and we’re actively planning around it. And it’s not due to shortages of our components—it’s also because of other critical elements that must be put in place, particularly around power and broad infrastructure connectivity globally, or at least across America, enabling inference workloads to be distributed to consumers and enterprises nationwide. So, we’re simply getting a lot of lead time.

Joe Moore

Very helpful. Thank you.

Operator

Thank you. We do have time for one final question, and that will come from the line of Joshua Buchalter with TD Cowen. Your line is open.

Joshua Buchalter

Hey guys, thank you for taking my question. In the past, you've talked about roughly $15–$20 billion per gigawatt of compute, and given the 10 gigawatts you’ve implicitly indicated you’ll be doing next year, that implies a number significantly larger than $100 billion. You’ve also mentioned that the value per gigawatt varies by project. So, how should we think about the evolution of your revenue per gigawatt over time? On one hand, I would expect average selling prices to increase on programs you're already shipping, but there are also other projects entering the model. Thank you.

Hawk

Our revenue—our content per gigawatt—will increase, driven simply by the fact that our compute chip, the XPU, will see a very dramatic price increase, particularly as you not only integrate SRAMs into it but also start embedding significant CPU functionality into the same XPUs, effectively making those chips multi-die packages with substantial HBM. As a result, the trajectory of content per gigawatt will continue to rise. It won’t necessarily increase every month, every six months, or even every quarter, but it will grow consistently from one generation to the next.

Joshua Buchalter

Thank you.

Operator

Thank you. I would now like to turn the call over to Jay, Head of Investor Relations, for closing remarks.

Jay

Thank you, Operator. Broadcom currently plans to report its earnings for the third quarter of fiscal year 2026 after close of market on Wednesday, September 2nd, 2026. A public webcast of Broadcom's earnings conference call will follow at 2:00 PM Pacific Time. That will conclude our earnings call today. Thank you all for joining. You may end the call.

Operator

This concludes today's program. Thank you all for participating. You may now disconnect.

Details at Broadcom IR

Tips: This summary was generated by an AI language model based on publicly available information, including auto-generated subtitles and publicly released audio from third-party sources. While efforts have been made to ensure accuracy, AI-generated content may contain errors or omissions. Please refer to the official materials released by the respective listed companies for the most accurate and up-to-date information. This content is for informational purposes only and does not constitute investment advice or a recommendation to buy or sell any investment products. The views and summaries presented do not reflect the official position of moomoo. Moomoo makes no warranties, express or implied, regarding the accuracy, timeliness, or completeness of this content and shall not be held liable for any losses arising from its use. The copyright of any original conference call content remains with the respective listed companies.

1.4M Views