Financial information provided by Financial Modeling Prep. Options data provided by Unusual Whales. Live transcripts provided by Quartr. Reported revenue and EPS data from Earnings powered by FinChat.io. All data is provided for informational purposes only, and is not intended for trading purposes or financial, investment, tax, legal, accounting or other advice.
In addition to U. S. GAAP reporting, Broadcom reports certain financial measures on a non GAAP basis. A reconciliation between GAAP and non GAAP measures is included in the table attached to today's press release. Comments made during today's call primarily refer to our non GAAP financial results. I will now turn the call over to Hock. Thank you, Gee.
13m 41s
And thank you everyone for joining us today. In our fiscal Q2 twenty twenty five, total revenue was a record $15,000,000,000 up 20% year on year. This 20% year on year growth was all organic. As Q2 last year was the first full quarter with VMware. Now revenue was driven. By continued strength in AI semiconductors and the momentum we have achieved in VMware. Now reflecting excellent operating leverage, Q2 consolidated adjusted EBITDA was $10,000,000,000 up 35% year on year. Now let me provide more color. Q2 Semiconductor revenue was $8,400,000,000 with growth accelerating to 17% year on year. Up from 11% in Q1. And of course, driving this growth was AI semiconductor revenue of over $4,400,000,000 which was up 46% year on year and continues the trajectory of nine consecutive quarters of strong growth. Within this, custom AI accelerators grew double digits year on year while AI networking grew over 170% year on year. AI networking. Which is based on Ethernet, was robust and represented 40% of our AI revenue. As a standards based open protocol, Ethernet enables one single fabric for both scale out and scale out. And remains the preferred choice by our hyperscale customers. Our networking portfolio of Tomahawk switches Jericho routers, and NICs is what's driving our success within AI clusters in hyperscalers.
16m 23s
And the momentum continues with our breakthrough Tomahawk six switch just announced this week. This represents the next generation 102.4 terabits per second switch capacity. TomOff six enables clusters of more than 100,000 AI accelerators to be deployed in just two tiers instead of three. This flattening of the AI cluster is huge. Because it enables much better performance in training next generation frontier models through a lower latency higher bandwidth, and lower power. Turning to XPUs or customer accelerators. We continue to make excellent progress on the multiyear journey of enabling our three customers and four prospects to deploy custom AI accelerators. As we had articulated over six months ago, we eventually expect at least three customers to each deploy 1,000,000 AI accelerated clusters in 2027 largely for training their frontier models. And we forecast and continue to do so a significant percentage of these deployments to be custom XPUs. These partners are still unwavering in their plan to invest despite these certain economic environment. In fact, what we've seen recently is that they are doubling down. On inference, in order to monetize their platforms. And reflecting this, we may actually see an acceleration of XPO demand into the back half of twenty twenty six.
18m 48s
To meet urgent demand for inference on top of the demand we have indicated from training. And accordingly, we do anticipate now our fiscal twenty twenty five growth rate of AI Semiconductor revenue to sustain into fiscal twenty twenty six. Turning to our Q3 outlook. As we continue our current trajectory of growth, We forecast AI semiconductor revenue to be $5,100,000,000 up 60% year on year which would be the tenth consecutive quarter of growth. Now turning to non AI semiconductors in Q2. Revenue of $4,000,000,000 was down 5% year on year. Non AI semiconductor revenue is close to the bottom has been relatively slow to recover. But there are bright spots. In Q2, [indiscernible] Enterprise Networking and Service Storage revenues were up sequentially. However, industrial was down [indiscernible] as expected, wireless was also down due to seasonality. In Q3, we expect Enterprise Networking and Broadband to continue to grow sequentially but server storage wireless, and industrial are expected to be largely flat. And overall, we forecast non AI semiconductor revenue to stay around $4,000,000,000 Now let me talk about our Infrastructure Software segment. Q2 Infrastructure Software revenue of 6,600,000,000.0 was up 25% year on year above our outlook of $6,500,000,000 As we have said before, this growth reflects our success in converting our enterprise customers from perpetual vSphere to the full VCF software stack subscription.
21m 25s
Customers are increasingly turning to VCF to create a modernized private cloud on prem, which will enable the to repatriate workloads from public clouds. While being able to run modern container based applications and AI applications. Of our 10,000 largest customers, over 87% have now adopted VCF. The momentum from strong VCF sales over the past eighteen months since the acquisition of VMware has created annual recurring revenue [indiscernible] otherwise known as [indiscernible] growth of double digits in a core infrastructure software.
22m 18s
In Q3, we expect infrastructure software revenue to be approximately $6,700,000,000 up 16% year on year. So in total, we are guiding Q3 consolidated revenue to be approximately $15,800,000,000 up 21% year on year. We expect Q3 adjusted EBITDA to be at least 66%. With that, let me turn the call over to Kirsten.
22m 56s
Thank you, Hock.
22m 58s
Let me now provide additional detail on our Q2 financial performance. Consolidated revenue was a record $15,000,000,000 for the quarter up 20% from a year ago. Gross margin was 79.4% of revenue in the quarter, better than we originally guided on product mix. Consolidated operating expenses were 2,100,000,000.0 of which $1,500,000,000 was related to R and D. Q2 operating income of $9,800,000,000 was up 37% from a year ago, with operating margin at 65% of revenue. Adjusted EBITDA was $10,000,000,000 or 67% of revenue, above our guidance of 66%. This figure excludes $142,000,000 of depreciation. Now, I'll review of the P and L for our two segments. Starting with semiconductors. Revenue for our semiconductor solutions segment was 8,400,000,000.0 with growth accelerating to 17% year on year, driven by AI. Semiconductor revenue represented 56% of total revenue, in the quarter. Gross margin for our Semiconductor Solutions segment was approximately 69% up 140 basis points year on year, driven by product mix. Operating expenses increased 12% year on year, to 971,000,000 on increased investment in R and D for leading edge AI semiconductors. Semiconductor operating margin of 57% was up 200 basis points year on year. Now moving on to infrastructure software.
24m 56s
Revenue for infrastructure software of $6,600,000,000 was up 25% year on year and represented 44% of total revenue.
25m 7s
Gross margin for infrastructure software was 93% in the quarter compared to 88% a year ago. Operating expenses were $1,100,000,000 in the quarter, resulting in infrastructure software operating margin of approximately 76%. This compares to operating margin of 60% a year ago. This year on year improvement reflects our disciplined integration of VMware. Moving on to cash flow. Free cash flow in the quarter was 6,400,000,000.0 and represented 43% of revenue. Free cash flow as a percentage of revenue continues to be by increased interest expense from debt related to the VMware acquisition, and increased cash taxes. We spent 144,000,000 on capital expenditures. Day sales outstanding were thirty four days in the second quarter, compared to forty days a year ago, We ended the second quarter with inventory of $2,000,000,000 up 6% sequentially in anticipation of revenue growth in future quarters. Our days of inventory on hand were sixty nine days in Q2, as we continue to remain disciplined on how we manage inventory across the ecosystem. We ended the second quarter with $9,500,000,000 of cash and $69,400,000,000 of gross principal debt. Subsequent to quarter end, we repaid $1,600,000,000 of debt, resulting in gross principal debt of 67,800,000,000.0 The weighted average coupon rate and years to maturity of our $59,800,000,000 in fixed rate debt is three point eight percent and seven years respectively.
26m 55s
The weighted average interest rate and years to maturity of our $8,000,000,000 in floating rate debt is five point three percent and two point six years. Respectively. Turning to capital allocation, In q two, we paid stockholders 2,800,000,000.0 of cash dividends based on a quarterly common stock cash dividend of $0.59 per share. In q two, we repurchased $4,200,000,000 or approximately 25,000,000 shares of common stock. In Q3, we expect the non GAAP diluted share count to be approximately 4,970,000,000.00 shares excluding the potential impact of any share repurchases. Now moving on to guidance. Our guidance for q three is for consolidated revenue of 15,800,000,000.0 up 21% year on year. We forecast semiconductor revenue of approximately 9,100,000,000.0 up 25% year on year.
27m 57s
Within this, we expect Q3 AI semiconductor revenue of 5,100,000,000.0 up 60% year on year.
28m 6s
We expect infrastructure software revenue of approximately 6,700,000,000.0 up 16%. Year on year. For modeling purposes, we expect Q3 consolidated gross margin to be down 130 basis points sequentially primarily reflecting a higher mix of XPUs within AI revenue. As a reminder, consolidated gross margins through the year will be impacted by the revenue mix of infrastructure software and semiconductors. We expect Q3 adjusted EBITDA to be at least 66%. We expect the non GAAP tax rate for Q3 and fiscal year twenty twenty five to remain at 14%, And with this, that concludes my prepared remarks. Operator, please open up the call for questions.
29m 4s
To time restraints, we ask that you please limit yourself to one question. Please stand by while we compile the Q and A roster. And our first question will come from the line of Ross Seymore with Deutsche Bank. Your line is open.
29m 21s
Hi, guys. Thanks for letting me ask a question. Hock, I wanted to jump onto the AI side in some of the commentary you had about next year. Can you just give a little bit more color on the inference commentary you gave? And is it more the XPU side, the connectivity side or both that's giving you the confidence to talk talk about the growth rate that you have this year being matched next fiscal year?
29m 43s
Thank you, Ross. Good question. I think we're indicating that what we are seeing and what we have quite a bit of visibility increasingly is [indiscernible] deployment of XPUs, next year and much more than we originally thought. And hand in hand, we did of course, more, more and more networking. So it's a combination of both.
30m 12s
In the imprint side of things?
30m 17s
Yeah. We're seeing much more inference now.
30m 22s
Thank you.
30m 24s
Thank you. One moment for our next question. And that will come from the line of Harlan Sur with JPMorgan. Your line is open.
30m 35s
Good afternoon. Thanks for taking my question and great job on the quarterly execution. Good to see the positive growth inflection quarter over quarter. Year over year growth rates in your AI business. You as a team, has mentioned, right, the quarters can be a bit lumpy. So if I smooth out kind of first 360% year over year. It's kind of right in line with your three year kind of SAM growth CAGR. Right? Your prepared remarks and knowing that your lead times remain at thirty five weeks or better, do you see the Broadcom team sustaining the 60% year over year growth rate exiting this year? And I assume that potentially implies that you see your AI business sustaining is 60% year over year growth rate into fiscal twenty six, again, based on your prepared commentary? Which again is in line with your SAM growth taker. Is that kind of the a fair way to think about the trajectory this year and next year?
31m 34s
Yeah. Holland, that's a very insightful set of, analysis here, and that's exactly what we're trying to do here, because over six months ago, we gave you guys a point a year 02/1927. As we come into the second now into the second half of 2025, and with improved visibility and updates we are seeing in the way our hyperscale partners are deploying data centers, AI classes. We are providing you more some level of guidance, visibility, what we are seeing how the trajectory of '26 might look like. I'm not giving you any update on '27. We're just still establishing the update we have in twenty seven six months ago. But what we're doing now is giving you more visibility into where we're seeing '26 headed.
32m 40s
But is the framework that you laid out for us, like, second half of last year, which implies 60% kind of growth CAGR in your SAM opportunity. Is that kind of the right way to think about it as it relates to the profile of growth in your business this year and next year?
32m 56s
Yes.
32m 58s
Okay. Thank you, Hock.
33m 2s
Thank you. One moment for our next question. And that will come from the line of Ben Reitzis with Melius Research. Your line is open.
33m 13s
Hey, how are doing? Thanks, guys. Hey, Hock, networking AI networking was really strong in the quarter. And it seemed like it must have beat expectations. I was wondering if you could just talk about the networking in particular, what caused that And how much of that is your acceleration into next year And when do you think you see Tomahawk kicking in? As part of that acceleration? Thanks.
33m 44s
Well, I I think the network AI networking as you probably would know, goes pretty hand in hand with deployment of AI accelerated clusters. It isn't it doesn't deploy, on the timetable that's very different from the way that accelerators get deployed, whether they are XPUs or GPUs. It does happen. And they deploy a lot in scale out where Ethernet, of course, is the choice of protocol, but it's also increasingly moving into the space of what we all call scale up within those data centers. Where you have much higher, more than we originally thought consumption or density of switches then you have in the scale out scenario. It's in fact, the increased density in scale up is five to 10 times more than in scale out. That's the part that kind of pleasantly surprised us, and which is why this past quarter, Q2, the AI networking portion continues at about 40% from when we reported a quarter ago for Q1. And, at that time, I said, I expect it to drop It has And and your thoughts on Tomahawk driving acceleration for next year and when it kicks in? Oh, six.
35m 26s
Oh, yeah. That's extremely strong interest now. We're not shipping big orders or any orders other than basic proof of concepts out to customers. But there is tremendous demand for this new 102 terabit per second Tomahawk switches.
35m 52s
Thanks, Hock.
35m 55s
Thank you. One moment for our next question. That will come from the line of Blaine Curtis with Jefferies. Your line is open.
36m 3s
Hey, thanks and great results. I just wanted to ask maybe following up on the scale out opportunity. So today, I guess your main is not really using an EnvyLink switch style scale up. I'm just kind of curious your visibility or the timing in terms of when you might be shipping a switched Ethernet scale up network to your customers?
36m 26s
The talking scale up?
36m 28s
Scale up.
36m 29s
Scale up.
36m 30s
Yeah. Well, scale up is very rapidly converting to Ethernet now. Very much so. It's I for our fairly narrow band of hyperscale customers, scale up is very much Ethernet.
36m 56s
Thank you. One moment for our next question. And that will come from the line of Stacy Rasgon with Burn Your line is open.
37m 7s
Guys, thanks for taking my questions. Hock, I still wanted to follow-up on that AI 2026 question. I want to just put some numbers on it. Just to make sure I've got it right. So if you did 60% in the first 360% year over year in q four, puts you at, like, I don't know, 5,800,000,000.0, something like 19,000,000,000 or 20,000,000,000 for the year. And then are you at are you saying you're gonna grow 60% in 2026, which would put you 30,000,000,000 plus in AI revenues for '26. I just want to make is that the math that you're trying to communicate to us directly?
37m 42s
I think you're doing the math. I'm giving you the trend. But I did answer that question, I think Harlan asked earlier. The rate we are seeing the now so far in fiscal twenty five, and will presumably continue. We don't see any reason why doesn't give an lead time visibility in '25. What we're seeing today based on what we have visibility on '26 is to be able to ramp up this AI revenue in the same trajectory. Yes.
38m 22s
So is is the cam going up So is the SAM going up as well? Because now you have inference on top of training. So is the SAM SAM still 60 to 90, or is the SAM higher now as you see it?
38m 34s
I don't I'm not playing the same game here. I'm just giving a trajectory towards where we drew the line on 27 before. So I have no no response to is the same going up or not. Stop talking about same now. Thanks.
38m 51s
Oh, okay.
38m 52s
Thank you.
38m 54s
One moment for our next question.
38m 57s
And that will come from the line of Vivek Arya with Bank of America. Your line is open.
39m 4s
Thanks for taking my question. I had a near and then longer term question on the XDU business. So, Hock, for near term, if your networking upsided in in q two, and overall AI was in line, it means XCU was perhaps not as strong. I realize it's lumpy, but anything more to read into that any product transition or anything else? So just a clarification there. And then longer term, you know, you have outlined number of additional customers that you're working with what milestones should we look forward to and what milestones are you watching to give you the confidence that you can now start adding that addressable opportunity into your 27 or 28 or other numbers? How do we get the confidence that these projects are going to turn into revenue in some, you know, reasonable timeframe from now? Thank you.
39m 56s
Okay, on the first part that you're asking, it's you know, it's like you're trying to be you're trying to count how many angels on a head of opinion. I mean, whether it's x XP or or networking. Networking is hard. But that doesn't mean experience any source. [indiscernible] It's very much along the trajectory we expect it to be. And there's no lumpiness, there's no softening. It's pretty much what we expect. The trajectory to go so far. And into next quarter as well, And probably beyond. So we have a fair it's a fairly, I guess, in our view, fairly clear visibility on the short term trajectory. In terms of going on to '27, no, we are not updating any numbers here. We, six months ago, we drew a sense for the size of the SAM based on you know, million x p million GPU XPU clusters for three customers, and that's still very valid at that point. That you'll be there. But and we have not provided any further updates here. Nor are we intending to at this point. When we get a better visibility clearer sense of where we are, and that probably won't happen until twenty six.
41m 23s
We'll be happy to give an update to the audience. But right now, though, in today's prepared remarks and answering a couple of questions, We have we are as we are doing, as we have done here, we are intending to give you guys more visibility what we've seen the growth trajectory in 'twenty six.
41m 53s
Thank you. One moment for our next question. And that will come from the line of C. J. Muse with Cantor Fitzgerald. Your line is open.
42m 3s
Yeah, good afternoon. Thank you for taking the question.
42m 6s
I was hoping to follow-up on Ross's question regarding inference opportunity. Can you discuss workloads that are optimal that you're seeing for custom silicon? And that over time, what percentage of your XPU business could be inference versus training? Thank you.
42m 21s
I think there's no differentiation between training and inference in using merchant accelerators versus customer accelerators. I think that all under the the whole premise behind going towards custom accelerators [indiscernible] Which is [indiscernible] cost alone. It is that as custom accelerators get used and get developed on a road map with any particular hyperscaler That's a learning curve. Learning curve on how they could optimize the way the algorithms on their large language models, gets written [indiscernible] typed to silicon.
43m 20s
And that ability to do so is a huge value added in creating algorithms that can drive their LLMs to higher and higher performance. Much more than basically segregation approach between hardware and the software. It says you literally combine end to end hardware and software as they take that. Take that journey. And it's a journey. They don't learn that in one year. Do it a few cycles get better and better at it. And then lies the value, the fundamental value in creating your own hardware versus using a third party merchant silicon. That you are able to optimize your software to the hardware and eventually achieve way high performance than you otherwise could. And we see that happening.
44m 29s
Thank you. One moment for our next question. And that will come from the line of Carl Ackerman with BNP Paribas. Your line is open.
44m 40s
Yes. Thank you. Hock, you spoke about the the much higher content opportunity in scale up networking I was hoping you could discuss how important is demand adoption for co package optics in in achieving this five to 10 x higher content for scale up networks. Or should we anticipate much of the scale up opportunity will be driven by Tomahawk and Thor and Nick's? Thank you.
45m 11s
I'm trying to decipher this question a yours. So let me try to answer perhaps in a way I think you want me to clarify.
45m 21s
First and foremost, I think most of what's scaling up there are a lot of the scaling up that's going in, or as I call it, means a lot of XPU or GPU to GPU interconnects. It's done on [indiscernible] Copper interconnects. And because, you know, the the size of this scale up cluster still not that huge yet, that you can get away with. Copper to using copper interconnects. And they're still doing it. Mostly, they're doing it today. At some point soon, I believe when you start trying to go beyond maybe 72 GPU to GPU, interconnects You may have to push towards a different protocol, by protocol mode at a different meeting. From copper to optical.
46m 26s
And when we do that, yeah, perhaps then things like exotic stuff like co packaging might be a fall off of, silicon with optical might become relevant. But truly, what we really are talking about is that at some stage, as the clusters get larger, which means scale up becomes much bigger you need to interconnect GPU or XPU to each other in scale up many more. Than just 72 or 100 maybe even 128, you start going more and more, You want to use optical interconnects simply because of distance. And that's when optical will start replacing copper. And when that happens, the question is what's the best way to deliver on optical. And one way is co packaged optics. But it's not the only way. You can just simply use continue use perhaps pluggable. At low cost [indiscernible] In which case then you can interconnect the bandwidth, the radix of a switch and our switch is down 512 connections. So you can now connect [indiscernible] [indiscernible] 512 for scale up phenomenon. And that was huge. But that's when you go to optical. That's going to happen within my view, a year or two. And we'll be right in the forefront of it And it may be co packaged optics, which we are very much in development but it's a lock in.
48m 13s
Co package or it could just be as a first step pluggable optic. Whatever it is, I think the bigger question is, when does it go from optical and from copper connecting GPU to GPU to optical. Connecting it. And the the step in that move will be huge. And it's not necessary for package updates, though that definitely one path we are pursuing.
48m 43s
Very clear. Thank you.
48m 45s
And one moment for our next question. And that will come from the line of Joshua Buchalter with T. Cowen. Your line is open.
48m 55s
Hey, guys. Thank you for taking my question. I realized it's a bit nitpicky, but I wanted to ask about gross margins in the guide. So your revenue implies sort of $800,000,000 incremental increase with gross profit up, I think, 400 to 450,000,000.
49m 9s
Which is kind of pretty well below corporate average fall through. Appreciate that semis are dilutive, and and custom is probably dilutive within semis but anything else going on with margins that we should be aware of? And and how should we think about the margin profile of custom longer term as that business continues to scale and diversify? Thank you.
49m 29s
Yeah, we've historically said that the XPU margins are slightly lower than the rest of the business. Other than wireless. And so, there's really nothing else going on other than that. It's just just exactly what I said, that the majority of it quarter over quarter is the 30 basis point decline is being driven by more XPUs. You know, there are more moving parts here. Than your simple analysis process. And I think your simple analysis is totally wrong in that regard.
50m 2s
Alright. And thank you.
50m 5s
And one moment for our next question. And that will come from the line of Timothy Arcuri with UBS. Your line is open.
50m 14s
Thanks a lot. I also wanted to ask about Scale Up, Hoc. So a lot of competing ecosystems. There's UA Link, which, of course, you left. And now there's the big, you know, GPU company, you know, opening up NVLink. And they're both trying to build eco systems. And there's an argument that you're an ecosystem of one. What would you say to that debate? Does does opening up NVLink change the landscape? Sort of how do you view of your AI networking growth next year? Do you think it's gonna to be primarily driven by scale up? Or would still be pretty scale out heavy?
50m 46s
It's I like to create platforms and new protocols and systems The fact of the matter is scale up. It can just be done easily, and it's currently available It's open standards, open source, Ethernet. Just as well just as well, don't need to create new systems for the sake of doing something that you could easily be doing in networking in Ethernet. And so, yeah, I hear a lot of this interesting new protocols standards that are trying to be created. And most of them, by the way, are proprietary. Much as they like to call it otherwise. One is really open source, and open standards is Ethernet. And we believe Ethernet will prevail as it does before for the last twenty years. In traditional networking. There are no reason to create a new standard for something that could be easily done in transferring bits and bytes of data.
51m 59s
Got it, Hak. Thank you.
52m 1s
And one moment for our next question. And that will come from the line of Christopher Rolland with Susquehanna. Your line is open.
52m 11s
Thanks for the question.
52m 13s
Yes, my question is for you, Hock. It's kind of a bigger picture one here. And this kind of acceleration that we're seeing in AI demand do you think that this is because of a marked improvement in ASICs or XPUs closing the gap on the software side at your customers? Do you think it's these require token [indiscernible] test time compute driving that, for example. What do you think is actually driving the upside here And do you think it leads to a market share shift faster than we were expecting towards XPU from GPU?
53m 3s
Thanks.
53m 5s
Yeah. Interesting question about now. None of the foregoing that you outlined. It's very simple, the way inference has come out. Very, very hot lately is remember, we're only selling to a few customers, hyperscalers with platforms and LLMs. That's it. They're not that many. And and you we told you how many we have. And haven't increased any. But what is happening is [indiscernible] and those with LNMs need to justify all the spending they're doing. Doing training makes your frontier model smarter. That's no question. It's almost like science. Research and science. Make your frontier models by creating very clever algorithm that that deep, consumes a lot of compute for training smarter. Training makes us smarter. You want to monetize inference.
54m 14s
And that's what's driving it. Monetize, I indicated in my prepared remarks, the drive to justify a return on investment and a lot of the investment is training, and then return on investment is by creating use cases a lot AI use cases, AI consumption, out there, through availability of a lot of the inference. And that's what we are now starting to see among a small group of [indiscernible] Excellent. Thank you.
54m 54s
And one moment for our next question. And that will come from the line of Vijay Rakesh with Mizuho. Your line is open.
55m 3s
Yeah, thanks. Hey, Hock. Just going back on the AI server revenue side, I know you said fiscal twenty five kind of tracking to that up 60% ish growth.
55m 14s
If you look at fiscal twenty six, you have many new customers lacking a meta and probably you have the four of the six hyperscalers that you're talking to past. Would you expect that growth to accelerate into fiscal twenty six? Evolve that [indiscernible] You know, my my my prepared remarks, which I [indiscernible] that the rate of growth we are seeing in 2025 will sustain into 2026 based on improved visibility and the fact that we're seeing inference coming in on top of the demand for training as the clusters get built up bigger and bigger, still stands. I don't think we are getting very far by trying to pass through my words or data here. It's just a Got it. [indiscernible] going from '25 into '26 as the best forecast we have at this point.
56m 11s
Got it. And and on the NBLINK the NVLink fusion versus the scale up do you expect that market to go the route of top of the rack where you've seen some move to the Ethernet side, in in in kind of scale out? Do you expect scale up to kind of go the same route?
56m 30s
Thanks. Well, Broadcom do not participate in NVLink. So I'm really not qualified to answer that question, I think.
56m 41s
Got it. Thank you.
56m 45s
Thank you. One moment for our next question. And that will come from the line of Erin Rakers with Wells Fargo. Your line is open.
56m 56s
Yes. Thanks for taking the question. I think all my questions on scale up have been asked. But I guess Hock, given the execution that you guys have been able to do with the VMware integration, looking at the balance sheet, looking at the debt structure, I'm curious if you could give us your thoughts on how the company thinks about capital return versus the thoughts on M and A the strategy going forward? Thank you.
57m 21s
Okay. That's an interesting question. And and I agree not too, not not too untimely, I would say. Because, yeah, we have done a lot of the integration of VMware now. You can see that in the the level of free cash flow we're generating from operations. And as we said, the use of capital has always been, we're very I guess, measured and upfront with a return through dividends, which is half our free cash flow of the preceding year. And frankly, as Kirsten has mentioned, three months ago and six months ago to in the last two earnings call, the first choice typically of the other free a part of the free cash flow to bring down our debt. To a more to a level that we feel closer to no more than two. Ratio ratio of debt to EBITDA. And that doesn't mean that opportunistically, we may go out there and buy back our shares. As we did last quarter. And indicated by Kirsten, we did $4,200,000,000 of stock buyback. Now part of it is used to basically when [indiscernible] [indiscernible] basically use we basically buy back part of the shares in used to be paying taxes on the invested ERSU But the other part of it, I do I do a main we use it opportunistically last quarter when we see an opportune situation when basically we think that it's a good time to buy some shares back.
59m 18s
We do. But having said all that, our use of cash outside the dividends would be, at this stage, used towards reducing our debt. And I know you're gonna ask what about M and A? Well, the kind of m and a we will do will, in our view, would be significant, would be substantial enough that we need [indiscernible] In any case. And a good use of our free cash flow to bring down debt to, in a way, expand, if not preserve our borrowing capacity if we have to do another M and A deal.
1h 2s
Thank you. One moment for our next question. That will come from the line of Srini Pajjuri with Raymond James. Your line is open.
1h 13s
Thank you. Hock, couple of clarifications. First, on your 2026 expectation, are you assuming any meaningful contribution from the four prospects that you talked about?
1h 30s
No comment.
1h 32s
We don't talk on prospects. We only talk on customers.
1h 36s
Okay. Fair enough. And then my other clarification is that I think you talked about networking being about 40% of the mix within AI. Is that the right kind of mix you expect going forward? Or is that going to materially change as we guess, see XPS ramping you know, going forward?
1h 56s
No. I've always say and I expect that to be the case in going forward in 2026 as we grow. [indiscernible] networking should be a ratio to XPU should be closer in the range of less than 30%. Not the 40%.
1h 1m 18s
Thank you. One moment for our next question. And that will come from the line of Joe Moore with Morgan Stanley. Your line is open.
1h 1m 29s
Great. Thank you.
1h 1m 31s
You've said you're not going to be impacted by export controls on AI. I I know there's been a number of changes since the industry since the last time you made the call. Is that still the case? And just know, can you give people comfort that you're there's no impact from that down the road?
1h 1m 46s
Nobody can give anybody comfort in this environment, Joe. You know that. Rules are changing quite dramatically as trade bilateral trade agreements continue to be negotiated in a very, very dynamic environment. So I'll be honest, I don't I don't know. I know as little as probably you probably know more than I do maybe. In which case then I know very little. About this whole thing about whether there's any export control, how the export control will take place we're guessing. So I'd rather not answer that because no, I don't know. Whether it will be.
1h 2m 30s
Thank you. And we do have time for one final question. And that will come from the line of William Stein with Truist Securities. Your line is open.
1h 2m 41s
Great. Thank you for squeezing me in. I wanted to ask about VMware. Can you comment as to how far along you are in the process of converting customers to the subscription model. Is that close to complete? Or is there still a number of quarters that we should expect that that that conversion continues?
1h 3m 5s
That's a good question. And so let me start off by saying a good way to measure it is you know, most most of our VMOA contracts are about three, typically three years. And that was what VMware did before we acquired them, and that's pretty much what we continue to do. Three is very traditional. So based on that, the renewals were like two thirds of the way, almost to the halfway more than halfway through the renewals. So we probably have at least another year plus, maybe a year and a half to go.
1h 3m 42s
Thank you. And with that, I'd like to turn the call over to GU for closing remarks. Thank you, operator. Broadcom currently plans to report its earnings for the third quarter of fiscal year twenty twenty five after close of market on Thursday, September 4, 2025. A public webcast of Broadcom's earnings conference call will follow at 2PM Pacific.
1h 4m 6s
That will conclude our earnings call today.
1h 4m 9s
Thank you all for joining. Operator, you may end the call.
Key Insights
Q2 FY25 total revenue reached a record $15B, up 20% YoY, driven by strong AI semiconductor growth and VMware momentum; adjusted EBITDA was $10B, up 35% YoY.
AI semiconductor revenue in Q2 was over $4.4B, up 46% YoY, with custom AI accelerators growing double digits and AI networking (Ethernet-based) up over 170% YoY, representing 40% of AI revenue.
Q3 FY25 guidance: consolidated revenue of ~$15.8B (up 21% YoY), AI semiconductor revenue of $5.1B (up 60% YoY), infrastructure software revenue of ~$6.7B (up 16% YoY), and adjusted EBITDA margin of at least 66%.
Management expects the FY25 AI semiconductor revenue growth rate (~60%) to sustain into FY26, citing increased demand for both training and inference workloads from hyperscale customers.
Infrastructure software (VMware) revenue was $6.6B in Q2 (up 25% YoY); over 87% of top 10,000 customers have adopted VCF, with the subscription conversion process expected to continue for another year to year and a half.
