Broadcom’s total AI-related orders on hand exceed $73 billion, nearly half of the company’s consolidated $162 billion backlog. The $73B backlog is expected to ship over the next 18 months. This backlog includes not only XPUs but also networking components. Most of the earnings call was management explaining the $73 billion is a baseline for the next 18 months.  

Notably, there’s been a significant amount of hype around custom silicon challenging Nvidia, thus, the bar was set high going into this earnings report. For Broadcom, the words “steady as you go” come to mind. 

Next quarter, AI revenue is expected to double year-over-year to $8.2 billion. During fiscal year 2025, AI revenue grew 65% year-over-year to $20 billion, leading to semiconductor revenue seeing an all-time high of $37 billion. During the fiscal year, the Infrastructure Software segment posted 26% growth to $27 billion, led by strong adoption of VMware Cloud Foundation, which represents enterprise software monetization. 

Management emphasized that AI has now grown more than 10x over the past 11 quarters, illustrating how rapidly Broadcom has scaled this business. Custom accelerators, or XPUs, more than doubled year-over-year, primarily driven by Google’s TPUs as Big Tech now turns toward monetizing their platforms through inference APIs and AI-driven applications. 

The report was fairly neutral as Broadcom struggled to live up to the recent custom silicon hype; yet it’s also clear Broadcom is in pole position to be a large beneficiary of the incoming AI Monetization Supercycle. You can view my free coverage here, where I connect the dots on why the AI trade’s best years are still up ahead. 

Regarding this earnings report, the main topics can summed up by the expanding customer list, the margin compression expected from XPUs, and the strength of Tomahawk 6. 

$73 Billion Visible Backlog and Expanding Customer List 

In the opening remarks, Hock Tan provided an update on the backlog, stating “And all these components combined with XPUs, bring our total order on hand in excess of $73 billion today, which is almost half Broadcom's consolidated backlog of $162 billion. We expect this $73 billion in AI backlog to be delivered over the next 18 months. And in Q1 fiscal '26, we expect our AI revenue to double year-on-year to $8.2 billion.” 

The earnings call was essentially a series of questions dissecting this statement. Overall, Tan implied this is more of a baseline, stating: “And obviously, this is as of now, I mean, we fully expect more bookings to come in over that period of time.” 

Broadcom has an expanding customer list that is quite impressive, including Google, Meta, Bytedance, Anthropic and now a fifth customer (analysts asserted the 5th customer is OpenAI, management declined to comment). The fourth customer, Anthropic, placed a $10 billion order for the TPU Ironwood racks with an additional $11 billion placed in the latest quarter. In the earnings call, management made sure to state they are building server racks for Anthropic and not only chips – stating it was “a system sale.”  

The market is suddenly taking notice of custom silicon (despite it being debated as a risk to Nvidia for over a decade) because an R&D lab is turning to TPUs and also now that Ironwood v7 is the first generation of TPUs to be specifically designed for inference.  

Tomahawk 6 

Broadcom’s Tomahawk 6 is an Ethernet switch built to address the scaling limits of AI clusters as they move beyond single-rack deployments by allowing hyperscalers to interconnect tens of thousands of accelerators with predictable performance, high bisection bandwidth, and tighter cost and power control. 

Tomahawk 6 delivers up to 102.4 Tbps of switching capacity and effectively doubles bandwidth versus the prior generation, enabling large-scale GPU and custom XPU fabrics to scale out while preserving low latency and power efficiency. Broadcom is making a bet that AI systems will increasingly rely on Ethernet for cluster expansion rather than proprietary fabrics (such as Nvidia’s NVLink).  

According to management, the new Ethernet switch is ramping quickly over the past 3 months and the current order backlog for AI switches exceeds $10 billion:

“And frankly, we see that bookings not just in XPUs, but in switches, DSPs, all the other components that go into AI data center. We have never seen bookings of the nature that what we have seen over the past 3 months, particularly with respect to Tomahawk 6 switches. This is one of the fastest-growing products in terms of deployment that we've ever seen of any switch products that we put out there. It is pretty interesting and partly because it's the only one of its kind out there at this point at 102 terabits per second. And that's that exact product needed to expand the clusters of the latest GPU and XPUs out there.” 

XPUs will Lead to Margin Compression 

If I were to point to why there is weakness after hours, it’s likely a combination of the $73 billion not meeting the high bar the custom silicon hype set for the company, but also the discussions around XPUs leading to margin compression over time.  

The company will have to pass-through more third-party components such as memory, optics, and power infrastructure, which will lead to gross margins contracting. However, management was clear that gross profit dollars and operating income dollars will continue to rise due to scale and operating leverage.  

According to the CFO: “And so those gross margins will be lower. However, overall, the way Hock said it, gross margin dollars will go up, margins will go down, operating margins — because we have leverage operating margin dollars will go up, but the margin itself as a percentage of revenues will come down a bit.” 

Financials 

Revenue grew by 28% 

Broadcom’s FQ4 ending October 2025 revenue grew by 28.2% YoY and 12.9% QoQ to $18.02 billion, beating estimates by 3.2%. Revenue growth accelerated by 6.2 percentage points from 22% growth reported in FQ3. The strong growth was primarily driven by a surge in AI revenue and growth in Infrastructure software revenue.  

Management also provided a strong FQ1 revenue guide of $19.1 billion, implying a YoY growth of 28.1% and 6% QoQ, beating estimates by 4.3%. The expected strong growth is primarily driven by AI revenue, which is expected to double YoY to $8.2 billion. Analysts expect strong growth to continue, with revenue expected to grow 26% YoY to $18.91 billion in FQ2 and accelerating 49.3% YoY growth to $23.82 billion in FQ3. 

For FY2025, ending October, revenue grew by 23.9% YoY to a record $63.89 billion. The strong growth was primarily driven by AI revenue and VMware. Looking forward, analysts expect revenue to grow 35.7% YoY to $86.1 billion in FY2026 and 33.1% YoY to $114.59 billion in FY2027. 

Key Segments 

Semiconductor Solutions 

FQ4 semiconductor solutions revenue grew by 35% YoY to $11.07 billion, primarily driven by strong AI revenue. Revenue growth accelerated by 9 percentage points from 26% growth reported in FQ3. Management expects semiconductor revenue growth to further accelerate 15 percentage points to 50% YoY, reaching $12.3 billion in FQ1, driven by a surge in AI revenue. For FY2025, semiconductor revenue grew by 22% YoY to a record $36.9 billion.  

FQ4 AI revenue grew by 74% YoY and 25% QoQ to $6.5 billion and was higher than the management guide of $6.2 billion. CEO Hock Tan said in the earnings call, “And this represents a growth trajectory exceeding 10x over the 11 quarters we have reported this line of business. Our custom accelerated business more than doubled year-over-year, as we see our customers increase adoption of XPUs, as we call those custom accelerators in training their LLM and monetizing their platforms through inferencing APIs and applications.” It further highlights the point that we have discussed in our article here that Broadcom is a silent beneficiary of the AI Monetization trend.  

Management also highlighted that these XPUs have also been extended to other LLMs, “best exemplified at Google, where the TPUs use in creating Gemini, have also been used for AI cloud computing by Apple, Coherent and SSI as an example. And the scale at which we see this happening could be significant.” Management confirmed that the $10 billion order from the fourth customer they mentioned in the last earnings call was from Anthropic and that they received an additional $11 billion order this quarter for delivery in late 2026. Broadcom also announced a fifth XPU customer this quarter, who has placed a $1 billion order to be delivered in late 2026. 

Management also provided a strong AI revenue guide for FQ1 of $8.2 billion, implying a 100% YoY and 26% QoQ growth. The expected strong growth is primarily driven by custom AI accelerators and Ethernet AI switches. For the FY2025, AI revenue grew by 65% YoY to $20 billion. Management expects AI revenue to accelerate in FY2026 and drive most of Broadcom’s growth in FY2026.

Non-AI semiconductor revenue in FQ4 grew by 2% YoY and 16% QoQ to $4.6 billion primarily driven by favorable wireless seasonality. As seen below, the gap between AI and non-AI revenue is widening as AI growth accelerates. Management expects non-AI-semiconductor revenue to be flat YoY to $4.1 billion and down sequentially in FQ1 due to wireless seasonality.  

Infrastructure Software 

FQ4 Infrastructure software revenue grew by 19% YoY to $6.9 billion, above the management guide of $6.7 billion. Bookings continue to be strong, with total contract value booked in FQ4 exceeding $10.4 billion compared to $8.2 billion in the same period last year. 

The Infrastructure Software backlog was $73 billion compared to $49 billion in the same period last year. Management expects renewals to be seasonal in Q1 and expects Infrastructure Software revenue to be $6.8 billion, down (2%) sequentially and up 1% YoY.  

For the FY2025, Infrastructure Software revenue grew by 26% YoY to $27 billion, primarily driven by strong VMware revenue. Management expects Infrastructure Software revenue to grow in the low double digits in FY2026. 

Margins 

Broadcom reported better margins than expected, primarily due to higher software revenue than expected, operating leverage, and better product mix within the semiconductor revenue. As discussed earlier in our article that AI revenue will lead to lower gross margin in the coming quarters. However, management was clear that gross profit dollars and operating income dollars will continue to rise due to scale and operating leverage.   

• FQ4 gross profits grew by 36.1% YoY to $12.25 billion, with a gross margin of 68%, an improvement of 390 basis points YoY and 90 basis points sequentially. Adjusted gross margin was 77.9%, up 100 basis points YoY and down 50 basis points sequentially. It was better than the management guidance of 77.7% primarily due to higher software revenues than expected and better product mix within semiconductors. Management expects FQ1 adjusted gross margin to be down 100 basis points sequentially to 76.9% primarily due to higher mix of AI revenue. 
• FQ4 operating income grew by 62.3% YoY to $7.5 billion. Operating margin improved 8.8 percentage points YoY and 4.8 percentage points sequentially to 41.7%, primarily driven by operating leverage. The adjusted operating margin was 66.2%, compared to 62.7% in the same period last year and 65.5% in the previous quarter. 
• Net income grew by 102.6% YoY to $8.5 billion with net profit margin of 47.3% compared to 30.8% in the same period last year. Adjusted net income grew by 39.5% YoY to $9.7 billion, with an adjusted net profit margin of 53.9% compared to 49.6% in the same period last year. 

FQ4 adjusted EBITDA grew by 34.4% YoY to $12.2 billion with an adjusted EBITDA margin of 68% and was better than the management guide of 67%. For FQ1, management expects adjusted EBITDA margin to be down 100 basis points sequentially and YoY to 67%.

• For FY2025 gross margins came at 67.8%, an improvement of 480 basis points YoY. Similarly, operating margin improved by 13.8 percentage points to 39.9%. The adjusted EBITDA margin was 67% compared to 62% last year.

Adjusted EPS grew by 37% 

FQ4 GAAP EPS grew by 93.3% YoY to $1.74. While adjusted EPS grew by 37.3% YoY to $1.95, beating estimates by 4.3%. Analysts expect adjusted EPS to grow by 23.3% YoY to $1.97 in FQ1 and 28.7% YoY to $2.03 in FQ2.  

Strong adjusted EPS is expected to continue in the coming years and analysts expect FY2026 adjusted EPS to grow by 39.1% YoY to $9.39 and 35.6% YoY to $12.72 in FY2027. However, these estimates are conservative, as the ramp-up of recent deals is expected to provide a further boost to the bottom line in the long term. 

Cash Flow and Balance Sheet 

Broadcom’s cash flows are improving, driven by higher profits. 

• FQ4 operating cash flows grew by 37.5% YoY to $7.70 billion with an operating cash flow margin of 42.8% compared to 39.9% in the same period last year. 
• FQ4 free cash flows grew by 36.2% YoY to $7.47 billion with a free cash flow margin of 41.4% compared to 39% in the same period last year. 
• Cash was $16.18 billion at the end of FQ4 with debt of $65.1 billion compared to $10.7 billion cash and debt of $64.2 billion at the end of FQ3; cash increased due to higher free cash flows in the recent quarter. 
• Management also increased the quarterly dividend by 10% to $0.65 or $2.60 for FY2026. 
• Inventory grew by 4% sequentially to $2.3 billion in FQ4.

Conclusion: 

Broadcom provided a solid report with no red flags to speak of. The AI cycle is approaching an inflection point, as a technology long debated will finally begin to move toward monetization, which will be a defining moment for the markets. If I had to guess, after listening closely to the management teams on the front lines, we will see major progress on inference in 2026 with more economic impact in 2027-2028.  

That makes 2025 the AI crux as many companies are spending an ungodly amount on building AI infrastructure with little immediate return on investment. When revenue and profits begin to catch up to these investments, the impact could be significant. I believe Broadcom will have a front row seat for that moment.  

You can read previous discussions around Broadcom’s custom silicon opportunity and networking opportunity in the deep dive on the Networking/ASICs Giant, the analysis covering the $110B backlog, and also This Stock is Set to Surge from AI Inference.Networking/ASICs Giant, the analysis covering the $110B backlog, and also This Stock is Set to Surge from AI Inference.

I/O Fund Equity Analyst Royston Roche contributed to this analysis.

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 AVGO at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• Coherent: Indium Phosphide Capacity to Double, Data Center to Reaccelerate to 10% QoQ
• Credo Fiscal Q2: Revenue Surges as Reliability Wins in a Crowded Market
• Nvidia Q3: Largest QoQ Growth in 2 Years; Networking up 162%
• CoreWeave Q3: Timing Miss yet Backlog up 2X QoQ and up 4X YTD

Nvidia’s Q3 showed the company’s GPU momentum return, delivering a substantial data center beat with 25% QoQ growth, surpassing an important $50 billion quarterly revenue milestone for the segment. More importantly, Nvidia’s guide pointed to this momentum continuing into the fourth quarter, implying that data center revenue could be on track to rise another $8 billion QoQ for 15% growth. 

Yesterday I published an article entitled Why Nvidia Stock Could Reach $20 Trillion Market Cap by 2030 – a prediction that requires a 36% CAGR over a five-year period or about 8% growth QoQ. These two quarters alone meet the criteria for next year’s CAGR plus some. 

Margins expanded sequentially, free cash flow sustained nicely, networking was up an impressive 162% and compute was up 56%. Management repeated they “currently have visibility to $0.5 trillion in Blackwell and Rubin revenue from the start of this year through the end of calendar year 2026.”

Below, we look at one other metric that hints that Blackwell Ultra can continue to deliver knockout earnings report while our sights are now set on Vera Rubin for H2 2026.  

Nvidia Surpasses $50 Billion Quarterly Data Center Revenue 

Nvidia surpassed the $50 billion quarterly data center revenue milestone in Q3, as it reported $51.2 billion in revenue for the segment, up 25% QoQ and 66% YoY. This is the highest QoQ growth rate for data center since fiscal Q4 2024. An impressive feat to deliver such strong growth at this scale considering the segment was just $18.4 billion at the time. On a dollar basis, data center revenue rose by $10.1 billion sequentially.  

This sequential growth was driven by a strong inflection in Compute revenue, which surged 27% QoQ to $43 billion, its highest sequential growth rate since fiscal Q1 2025; however, this does come after a (1%) QoQ decline in fiscal Q2. Nvidia noted that Blackwell Ultra was ramping across all customer categories and became its leading architecture.  

Q4’s guidance suggests that this $50 billion data center segment will quickly be in the rear view mirror, with the $65 billion guidance implying data center revenue of around $59 billion assuming similar mix shift as Q3. This represents another 15% QoQ growth on top of Q3’s 25%, or essentially the data center segment rising nearly 44% in just two quarters.  

This would also correspond to a nearly $8 billion QoQ increase, meaning that if Nvidia maintains this growth cadence through mid-CY26, then it would reach our prediction for a $75 billion data center segment two quarters early. If this materializes, this would represent data center growth of 66% YoY, up from 56% last quarter.  

It also could suggest Nvidia potentially reaching a $90 billion quarterly data center segment if this trajectory is maintained through the end of fiscal 2027. However, it is important to note that given the sheer scale of data center revenue, there is the potential for this inflection to be lumpy. Therefore, we are directionally bullish but not with too specific of a timeline; rather, these are milestones that tell us how quickly we could see $8 trillion market cap for example and then onward. 

Blackwell Revenue Tops $100 Billion 

As stated in our $20 Trillion pre-earnings writeup, after taking into account Blackwell revenue that will ship in FY2026, this should lead to a $320 billion data center segment next year. Here is what was stated in the $20 Trillion analysis: “Reading between the lines on Huang’s comments suggests strong upside to Nvidia’s data center revenue through 2026. Over the prior three quarters heading into fiscal Q3’s report, Blackwell revenue has totaled approximately $63 billion. Including Networking over that time frame, total revenue would rise to $78 billion, still a fraction of the total overall opportunity management is projecting. Thus, if we assume that Blackwell and Rubin ramp over the next five quarters, fiscal 2027 data center revenue could be nearly $320 billion, versus estimates for around $270 billion.”  

Since we wrote that earlier this week, analyst estimates have been rising and now stand at $292 billion for next year. With information from this report, we look to be on track for a $317 billion data center segment next year (close to our original estimate this morning). 

We calculated this from the prior three quarters heading into fiscal Q3’s report, Blackwell revenue has totaled approximately $63 billion. Now, Q3’s Compute revenue of $43 billion implies Blackwell has delivered around $104 billion in revenue in the past four quarters, assuming the only non-Blackwell revenue was the $2 billion disclosed from Hopper.  

Including Networking and Q4’s guidance, Nvidia looks to be on track to generate $186 billion of its $500 billion opportunity in fiscal 2026. This would leave approximately $314 billion for fiscal 2027’s data center revenue to meet the $500 billion visibility, but if Nvidia can exceed that by 2-4%, it could be on track for $330 billion next year. Management sounded confident to achieve the $500 billion target and they hinted that they could exceed it as the CFO stated, “So there's definitely an opportunity for us to have more on top of the $500 billion that we announced.”

One Figure Says This Growth Inflection Will Continue 

While we continue to hammer on the importance of Big Tech’s capex as the number one indicator for Nvidia’s data center growth continuing, there was potentially a more important, well overlooked figure in Nvidia’s report that signals this data center inflection will continue.  

Nvidia’s total supply-related commitments, such as for CoWoS wafers, HBM memory, or other components, surged nearly 52% QoQ to $50.3 billion in Q3, with management noting that they are “ordering to secure long lead-time components, meet the demand for Blackwell, and support future architecture ramps.”  

This is a notable increase from the prior five-quarter average of ~$30 billion, which is likely supporting the current ramp in data center revenue. This uptick in supply commitments, which is likely to translate into inventories and revenue over the coming four to six quarters, hints that Nvidia will continue ramping Blackwell output while preparing for Rubin’s production in the second half of 2026.   

This also bolsters confidence in Nvidia’s order visibility to fill out and even exceed this cumulative $500 billion in Blackwell and Rubin revenue, as the company would not need to boost supply commitments by this degree if the demand signals were not there.  

There is Global Demand for Nvidia’s GPUs 

When hearing about an AI bubble, it’s important to remember there is global demand for Nvidia’s GPUs. The diversification across geographic regions, enterprises, startups – and of course, Big Tech, helps to insulate Nvidia should one customer or region slow their spending. Here is what was stated on the call: 

“And then lastly, remember, we were just talking about the American CSPs. Each country will fund their own infrastructure. And you have multiple countries, you have multiple industries. Most of the world's industries haven't really engaged agenetic AI yet, and they're about to. All the names of companies that you know we're working with, whether it's autonomous vehicle companies or digital twins for physical AI for factories and the number of factories and warehouses being built around the world, just a number of digital biology start-ups that are being funded so that we could accelerate drug discovery. All of those different industries are now getting engaged, and they're going to do their own fundraising. And so don't just look at the hyperscalers as a way to build out for the future. You got to look at the world, you got to look at all the different industries and enterprise computing is going to fund their own industry.” 

Commentary on Vera Rubin 

If only a Nvidia investor could kick back and call it a day! Instead, given Blackwell Ultra now comprises 2/3 of revenue confirming a successful launch, our sights are now set on Vera Rubin commentary. According to management, Rubin is set for a fast ramp: 

“The Rubin platform is on track to ramp in the second half of 2026. Powered by 7 chips, the Vera Rubin platform will once again deliver an X-factor improvement in performance relative to Blackwell. We have received silicon back from our supply chain partners and are happy to report that NVIDIA teams across the world are executing to bring up beautifully.  

Rubin is our third-generation rack-scale system substantially redefined the manufacturability while remaining compatible with Grace Blackwell. Our supply chain data center ecosystem and cloud partners have now mastered the build to installation process of NVIDIA's rack architecture. Our ecosystem will be ready for a fast Rubin ramp.” 

Financial Overview: 

Strong Revenue Growth of 63% 

Nvidia’s Q3 revenue grew by a solid 62.5% YoY and 22% QoQ to $57.01 billion. Revenue growth accelerated by 6.9 percentage points from 55.6% YoY growth reported in Q2. Revenue beat estimates by 3.5% and is the strongest beat in the last four quarters. The company’s strong revenue growth dispelled fears of an AI Bubble. Nvidia’s CEO Jensen Huang said, “Blackwell sales are off the charts, and cloud GPUs are sold out.”  

The Blackwell revenue gained further momentum in the recent quarter. The GB300 sales were higher than the GB200 sales, notably accounting for 2/3 Blackwell’s revenue, driven by strong demand from cloud companies and hyperscalers. The Hopper platform contributed approximately $2.0 billion in revenue. While H20 sales were negligible at $50 million, management is working with the US and Chinese governments to ship products to China. Looking forward, Rubin is on track to ramp in the second half of 2026. 

Management also provided a strong Q4 revenue guide of $65 billion at midpoint, representing a YoY growth of 65.3% and up 14% QoQ. It beat the estimates by 5.1%. Looking forward, analysts expect revenue to grow 40.5% YoY to $292.1 billion for FY2027 and 24.8% YoY to $364.6 billion for FY2028.

Networking Revenue Growth of 162% 

The company’s networking revenue was an outlier, growing 162% YoY and 13% QoQ to $8.19 billion. Revenue growth accelerated by 84 percentage points from 78% YoY growth in Q2. Management stated in the earnings call that the company’s networking business is specifically built for AI and is now the largest in the world. The strong growth was primarily due to NVLink scale-up and robust double-digit growth across Spectrum-X Ethernet and Quantum-X InfiniBand. 

Management stated in the earnings call that Meta, Microsoft, Oracle, and xAI are building gigawatt AI factories with Spectrum-X Ethernet switches, further highlighting the flexibility and openness of the company’s platform. 

The company introduced Spectrum-XGS Ethernet in August, which will enable to connect distributed data centers into Giga-Scale AI Super-Factories. Nvidia is the only company with AI scale-up, scale-out and scale across platforms, reinforcing the unique position in the market as the AI infrastructure provider. 

• Q3 gaming revenue grew by 30% YoY and was down (1%) sequentially to $4.27 billion. Management mentioned that channel inventories have reached more normalized levels heading into the holiday season. 
• Pro visualization revenue grew by 56% YoY and 26% sequentially to $760 million. Colette Kress, CFO, said in the earnings call, “Growth was driven by DGX Spark, the world's smallest AI supercomputer, built on a small configuration of Grace Blackwell.” 
• Automotive revenue grew by 32% YoY and up 1% QoQ to $592 million. The CFO highlighted, “We are partnering with Uber to scale the world's largest Level 4 ready autonomous fleet built on the new NVIDIA Hyperion L4 robotaxi reference architecture.” 
• OEM and other revenue grew by 79% YoY and 1% QoQ to $174 million.  

Margins 

The company’s margins beat management guidance and are expected to expand in Q4. 

• Q3 gross profits grew by 60% YoY to $41.85 billion. Q3 gross margin was 73.4%, beating management guidance of 73.3% by 10 basis points. Gross margin was up 100 basis points sequentially and down 120 basis points YoY. Q4 gross margin guide is 74.8%, up 140 basis points sequentially and up 180 basis points YoY.  Adjusted gross margin was 73.6%, beating the management guidance by 10 basis points. Management expects an adjusted gross margin of 75% in Q4, up 140 basis points sequentially and up 150 basis points YoY. 
• Looking forward, management mentioned in the earnings call that the input costs are increasing and are looking to hold gross margins in the mid-70s range for FY2027. 
• Q3 operating income grew by 65% YoY and 27% sequentially to $36.01 billion. The operating margin was 63.2%, beating the management guidance by 80 basis points. Adjusted operating margin was 66.2%, beating the management guidance by 50 basis points. Management has provided a strong operating margin guide of 64.5% and an adjusted operating margin guide of 67.3% for Q4. 
• Q3 net profits grew by 65% YoY and 21% QoQ to $31.9 billion with a net profit margin of 56% compared to 55% in the same period last year and 56.6% in Q2. Adjusted net profits grew by 59% YoY and 23% QoQ to $31.77 billion with an adjusted net profit margin of 55.7%, compared to 57% in the same period last year and 55.2% in Q2.

Adjusted EPS grew by 60.5% 

Q3 adjusted EPS grew by 60.5% YoY and 23.8% QoQ to $1.30, beating estimates by 3.5%. GAAP EPS grew by 66.7% YoY to $1.30, beating estimates by 8.5%. GAAP EPS included $0.06 in gains in non-marketable and publicly held equity securities. 

• Analysts expect adjusted EPS to grow 61.2% YoY to $1.43 in Q4 and accelerate to 89.5% YoY growth to $1.53 in Q1. 
• Looking forward, analysts expect FY2027 adjusted EPS to grow 49.5% YoY to $6.83 and 26.7% YoY to $8.65 in FY2028.

Cash and Balance Sheet 

The company has a strong balance sheet with solid cash flows primarily driven by strong revenue and profits.  

• Q3 operating cash flow grew by 34.7% YoY to $23.75 billion with an operating cash flow margin of 41.7%, compared to 50.3% in the same period last year and 32.8% in Q2. 
• Q3 free cash flows grew by 31.6% YoY to $22.09 billion with a free cash flow margin of 38.7%, compared to 47.9% in the same period last year and 28.8% in Q2. 
• The company’s cash and marketable securities have been steadily increasing and were $60.6 billion at the end of Q3, up from $56.8 billion in Q2 and $38.5 billion in the same period last year. Debt remained constant at $8.47 billion for Q3 and Q2. 
• The company returned $12.7 billion to shareholders in the third quarter through $12.5 billion of share repurchases and $243 million of cash dividends. The company expects to continue to use its strong future cash flows to buy back shares and invest in AI growth opportunities.  
• Inventories grew by 32% sequentially to $19.78 billion to support strong revenue growth. 

Conclusion: 

The $20 trillion prediction came from looking at the original $10 trillion prediction ahead of earnings and realizing I’d have to bump this up given the commentary around the $500 billion from the Blackwell-Rubin cycle. Although we had already slated next year for a $300 billion run rate and a $75 billion quarterly data center segment, it helped to hear last month that management agrees this is possible. Where the disconnect happens with analyst estimates is what will happen after next year as this is where analyst estimates show minimal growth through 2030 revenue with $437 billion whereas I am calling for double that by 2030. While Blackwell Ultra gets us to a new milestone of $50 billion to $75 billion quarterly revenue, quite a bit of my thesis depends on Vera Rubin, Rubin Ultra and the Feynman generations. 

Although the next five years will be a marathon, the Q3 report is a step in the right direction. I don’t expect consistent QoQ growth every quarter, yet as stated, we are already exceeding my CAGR for next year in two quarters’ time. Crunching these numbers matters quite a bit as we are talking about the world’s most valuable company and Nvidia will have to put up consistent growth for the stock to inch upward. The days of a sudden spike in the stock price are likely behind us, yet if Nvidia remains consistent, the incessant market narratives will eventually tire.  

Overall, this was an excellent report — and after two years of dissecting every angle of Blackwell, I’m excited to finally shift coverage to the Rubin generation of GPUs. Woohoo! This now marks the fourth GPU generation I’ve retired for I/O Fund Members — and we’re officially moving on to the fifth.

I/O Fund Equity Analysts Damien Robbins and Royston Roche contributed to this analysis.

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:

• The I/O Fund’s Top 15 AI Stocks for Q4 2025
• CoreWeave Q3: Timing Miss yet Backlog up 2X QoQ and up 4X YTD
• AMD Q3: The Catalyst is Expected in H2 2026, Could Ramp Sooner
• Reddit Q3: Setting a High Bar with Top Line Strength and 10-point Sequential Margin Expansion

Broadcom Q3 results confirm the stock as the number two player in AI as a combination of AI networking, custom silicon and AI software propelled total revenue to record levels. Semiconductor Solutions accelerated nine points to 26% YoY growth due to a rebound in AI accelerators (+63% YoY) and networking (170% YoY). While non-AI Semiconductor revenue remains weak and flat sequentially, VMware’s contribution helped prop up the segment.  

Looking ahead, Broadcom’s Q4 guide implies acceleration into year-end, with total revenue expected to reach $17.4B. More importantly, we picked up on a subtle shift in commentary from CEO Hock Tan regarding AI revenue growth in FY26, which should place Broadcom firmly above $30 billion for next fiscal year. Tan stated: “[…] we now expect the outlook for fiscal 2026 AI revenue to improve significantly from what we had indicated last quarter.”  

As we previously discussed this summer, demand for inference is booming. Broadcom’s edge goes beyond the fact that custom accelerators are often multiples cheaper than Nvidia’s GPUs for inference tasks – it's that custom silicon is increasingly performant with each generation. By optimizing algorithms (software), Big Tech can drive higher performance from large language models (LLMs) — which helps to drive down costs while also increasing output for specific workloads. For example, a rough idea as to how much it costs Nvidia to make merchant GPUs is estimated around $3,000 to $5,000 whereas the company charges $25,000 to $30,000 – hence the AI leader’s excellent margins. Reducing Nvidia’s high pricing power is what Big Tech is after and this can be accomplished both in the hardware costs but also through optimizing the workloads for specific use cases. 

Big Tech is prominent in Broadcom’s custom silicon customer list, which includes Google and Meta. ByteDance reportedly emerged as the third customer last summer. Tonight, a fourth customer was announced for a $10 billion XPU order, which was likely either OpenAI or perhaps Apple, as both were heavily rumored to be prospective customers (note: customer name was not offiically confirmed) 

“Last quarter, one of these prospects release production orders to Broadcom. And we have accordingly characterized them as a qualified customer for XPUs. And, in fact, has secured over $10 billion of orders of AI regs based on our XPUs.” 

Perhaps most interesting, Hock Tan slipped into the opening remarks that their backlog is at $110 billion. I was glad when an analyst circled back to this comment as this backlog is astonishing, to say the least.  

More details are below!

170% AI Networking Growth – Outpacing Street Models 

Q3’25 Revenue was $15.95 billion, beating estimates for $15.82 billion, and reflecting top line growth of 22.0% YoY and 6.3% QoQ. This represents record quarterly revenue, driven by custom AI accelerators, networking switches, and VMware strength in Infrastructure Software.  

Looking ahead, management provided Q4’25 guidance of $17.4 billion of revenue, implying 24% YoY growth and a slight uptick to 9% QoQ growth. This guide suggests that AI strength will more than offset flat non-AI revenue, while VMware remains a stable contributor. 

Key Segments 

Semiconductors were the main growth driver, fueled by AI accelerators and Ethernet networking. Infrastructure Software, which includes VMware acquisition, remained strong even though growth decelerated slightly when compared to prior quarters, 

Semiconductor Solutions Revenue accelerated nine points to 26% YoY at $9.17 billion. AI Semiconductor revenue surged 63% YoY to $5.2B, showing re-acceleration after a slower Q2 (+46% YoY). AI now represents 57% of Semiconductor revenue and 32% of total company revenue.  

Management guided Q4 AI revenue to $6.2B, which would represent ~19% sequential growth and eleven consecutive quarters of YoY growth. Though Broadcom did not lay out a FY25 AI revenue target, Q4 implies Broadcom is guiding for $19.9 billion in AI revenue for the year, up 63% YoY from $12.2 billion in FY24. 

Non-AI Semiconductor revenue remained flat at ~$4B, indicating continued end-market softness outside of AI. As seen below, the gap between AI and non-AI revenue is widening as AI growth accelerates. 

Infrastructure Software Revenue of $6.8B, ahead of guidance of $6.7B, up $190M QoQ from $6.6B in Q2’25 and up from $5.8B in Q3’24. $6.7B in revenue represents 43% of total revenue and 17% YoY Growth. 

Operating Leverage Offsets Gross Margin Compression 

Broadcom delivered balanced profitability in Q3 FY25. While AI hardware growth slightly compressed GAAP and Non-GAAP gross margins, the Company leveraged its scale and VMware’s recurring revenue to expand operating and net margins YoY. Broadcom’s ability to maintain a 67% EBITDA margin during a period of rapid AI expansion demonstrates exceptional cost control and pricing power, positioning it well for the Q4 acceleration. See below for in-depth breakdown of GAAP and Non-GAAP margin figures: 

GAAP gross margin of 67.1%, down 90 bps QoQ from 68% in Q2, and up from 63.9% in the same period last year. The margins were down sequentially due to custom silicon mix, which typically carries lower margins. Broadcom is flexing its ability to manage cost structures even as AI accelerators scale rapidly. 

Non-GAAP gross margin of 78.4%, down 100 bps QoQ from 79.4% in Q2 and up 100 bps YoY. This is driven by the increase in AI hardware revenue, which has lower gross margins compared to VMware’s high-margin software business. 

GAAP operating margin of 36.9%, down 190 bps QoQ from 38.8% in Q2, but up significantly from 29% in the same period last year. QoQ decline reflects seasonal opex and YoY improvement reflects strong operating leverage. 

Non-GAAP operating margin of 65.5%, slightly up QoQ from 65.3% in Q2’25 and up 180 bps from 63.7% in Q3’24. Broadcom is maintaining tight opex control even as it invests heavily in AI and networking. Non-GAAP margin stability reflects strong execution in managing both segments. 

GAAP Net Margin of 26.0%, down from 33.1% in Q2’25 and up from (14.3%) in Q3’24. Non-GAAP Net Margin of 52.7%, up from 51.9% in Q2’25 and up from 43.6% in Q3’24. Adjusted profitability is expanding as Broadcom benefits from both VMware’s recurring software revenue and AI-driven scale. 

Adjusted EBITDA of $10.7B, compared to $10.0B in Q2’25 and $8.2B in Q3’24. This represents an Adjusted EBITDA margin of 67.1%, compared to 66.7% in Q2’25 and 62.9% in Q3’24. 

Key Takeaways: 

• AI Mix Impact on GM: As AI Semi revenue grows (now sits at 32% of company revenue) there is slight gross margin pressure because custom silicon and hardware products carry lower margins. 
• Offset by Operating Leverage: Despite gross margin compression, non-GAAP operating margin expanded YoY thanks to scale and cost efficiencies. 
• Stable EBITDA Margin: Maintaining a 67% EBITDA margin while growing 22% YoY highlights Broadcom’s unique profitability profile when compared to peers. 
• Watch Forward Trend: With AI projected to reach $6.2B in Q4, mix shift could continue to pressure GAAP gross margins further. Look for growth from VMware and networking to keep non-GAAP margins steady or even up. 

EPS Growth driven by Top-Line Growth & Economies of Scale 

Non-GAAP EPS growth is the real story here: up 38% YoY, reflecting Broadcom’s ability to capitalize on AI strength with containing costs. GAAP EPS lagged slightly due to accounting for VMware’s amortization and elevated interest expense – not driven by any core business weakness.

EPS growth of 36.3% outpaced revenue growth of 22%, highlighting the company’s dual-engine model: AI hardware scale drives top-line growth while VMware’s recurring software offsets margin volatility and lifts EPS. Investors will appreciate the strong non-GAAP EPS beat and expansion in operating margins, especially given Broadcom’s premium valuation, where both EPS stability and growth are critical to sustaining multiples.  

• GAAP EPS of $1.02, compared to $1.03 in prior quarter and ($0.40) in prior-year quarter. 
• Non-GAAP EPS of $1.72, compared to $1.58 in prior quarter and $1.24 in prior-year quarter.  

Record Free Cash Flow Conversion 

Broadcom converted 44% of revenue into free cash flow, placing it among the top-tier semiconductor companies for cash generation. Strong OCF growth driven by AI semiconductor momentum and VMware’s software contribution. Broadcom continues to de-risk its balance sheet, reducing debt by $3B sequentially while building cash reserves and modestly increasing inventory ahead of anticipated Q4 AI revenue acceleration. With high FCF conversion and a strong balance sheet, Broadcom should have ample flexibility to return capital to shareholders and fund future AI growth initiatives, reinforcing its premium valuation. See below for an breakdown on cash flow figures: 

• GAAP operating cash flow of $7.2B, up 10% QoQ compared to $6.5B in Q2’25 and 57% YoY $4.6B in Q3’24. This represents an operating cash flow margin of 44.9%, up from 43.7% in Q2’25 and 36.7% in Q3’24. 
• Free Cash Flow of $7.0B, compared to $6.41B in Q2’25 and $4.5B in Q3’24. This represents a free cash flow margin of 44.0%, compared to 42.7% in Q2’25 and 35.6% in Q3’24. 
• Cash and Cash Equivalents of $10.7B, up from $9.5B as of Q2’25 and up from $10.0B as of Q3’24.  
• Debt of $64.2B, down from $67.2B in Q2’25, and down from $69.9B in Q3’24.  
• Inventory of $2.2B, up from $2.0B in Q2’25, and up from $1.8B in Q3’24. 

Valuation 

Broadcom currently trades at a premium to Nvidia, sitting at 19x forward revenue vs Nvidia 17.6x. Based on forward earnings, AVGO trades at 38x forward earnings, 13% above Nvidia and 18% above semi-industry average.  

This premium suggests that the market is pricing in 70%+ AI revenue CAGR through FY27, above management’s current 60%. The implication here: Any sign of AI growth slowing or continued margin pressure could trigger a valuation reset.  

Earnings Call Q&A 

Hock Tan Hints that Strong AI Growth is Incoming 

Last quarter, Hock Tan had stated the following: “And reflecting this, we may actually see an acceleration of XPU demand into the back half of 2026 to meet urgent demand for inference on top of the demand we have indicated from training. And accordingly, we do anticipate now our fiscal 2025 growth rate of AI semiconductor revenue to sustain into fiscal 2026." 

This evening, the following was stated:  

“we now expect the outlook for fiscal 2026 AI revenue to improve significantly from what we had indicated last quarter.” 

When asked later to clarify, Tan reaffirmed our understanding that it would mean above 60% growth and it was also stated the majority of the growth would come from XPUs with networking’s share of AI revenue declining next year due to XPU strength. 

“Let's answer the first part first, if I could be so bold as to suggest to you, when I — last quarter when they said, "Hey, the trend of growth of '26 will mirror that of '25which is 50%, 60% year-on-year. That's really all I said. I didn't — but of course, it comes up 50%, 60% because that's what '25 is.  

All I'm saying, if you want to put another way of looking at what I'm saying, which is perhaps more accurate is we're seeing — the growth rate accelerates as opposed to just remain steady at that 50%, 60%. We are expecting and seeing 2026 to accelerate more than the growth rate we see in '25. And I know you love me to throw in the number at you, but we are not supposed to be giving you a forecast for '26. But best way to describe it, it will be fairly material improvement.” 

Back of the napkin math:

Given that Hock Tan disclosed that Broadcom had secured $10B+ of orders related to the new fourth customer, current growth projections may materialize far too low. We have some estimates from HSBC placing just ASICs revenue at $28.3B (+128% YoY) with only $2.5B contribution from customers outside GOOG and META. 

The Street is at $19.9B for ASICs in FY26; assuming the Street is still at $28-29B AI revenue for FY26, this places Networking at $8-9B. 

However, assuming XPU strength and strong networking demand (as seen throughout the ecosystem) can drive an acceleration to 70% YoY in FY26, this projects $33.8B in AI revenue up from $19.9B estimated in FY25 given the implied guide. 

This ramp is supported by estimated ~80% increase in CoWoS allocation from ~83K in FY25 to ~150K in FY26, after Broadcom reportedly increased orders last week from 120-125K expected in FY26. 

Notably, Hock Tan did say the additional $10 billion would be recognized in Q3 of FY2026 during the call.  

$110 Billion Backlog 

Broadcom stating a $110 billion backlog was not on my Bingo card tonight. Wow! That is certainly a strong statement in terms of Broadcom’s prospects for continued AI growth. It’ll take me a day or two to process what that could mean for the next 1-3 years, assuming some of this is supply constrained.  

An analyst did ask for clarity and since it’s a rather large number to provide, especially since it’s primarily AI-driven, I’m quoting it in full for you below: 

Stacy Rasgon. Bernstein & Co. 

I was wondering if you could parse out this $110 billion backlog. Did I hear that number right? Could you give some color on the makeup of it — like how far does that go — and like how much of that $110 billion is AI versus non-AI versus software? 

Kirsten Spears   CFO: 

Well, yes, Stacy, we generally don't break up back on digital to give you a sense of how strong the business is as a whole for the company, and it's largely driven buying AI in terms of growth. Software continued to add on a steady basis. And non-AI, as I indicated, has grown double digits. Nothing compared to AI, which has grown very strongly. Give you a sense, perhaps fully 50% of it at least is semiconductors. 

Stacy Rasgon, Bernstein: 

Okay. And it's fair to say that semiconductor piece, it's going to be much more AI than non AI. 

Hock Tan, President: 

Right. 

Custom Silicon is Progressively Gaining Market Share 

During the discussion, an analyst asserted that custom silicon could surpass GPUs in terms of market share. That’s a tall order given Nvidia will typically be 1-2 years ahead of any custom programs (thereby offering an advantage to those who remain with their GPUs) and the two companies will likely end the year nearly $180B apart in AI revenue with AVGO around $20B and NVDA over $200B (could see $212B). 

However, I’ve also argued inference will provide an opening for Broadcom and AMD to meaningfully compete on AI accelerators. Therefore, I’m all ears and we will be watching this closely as we move along 2026-2028. 

Harsh Kumar, Piper Sandler: 

Hock, congratulations on all the exciting AI metrics and thanks for everything you do for Broadcom and sticking around. Hock, my question is, you've got 3 to 4 existing customers that are ramping. As the data centers for AI clusters get bigger and bigger, it makes sense to have differentiation, efficiency, et cetera, therefore, the case for XPUs. Why should I not think that your XPU share at these 3 or 4 customers that are existing will be bigger than the GPU share in the longer term? 

Hock Tan, CEO: 

It will be. It's a logical conclusion, Harsh, you're correct. And we are seeing that step by step. As I say, it's a journey. It's a multiyear journey because it's multigenerational, because these XPUs don't stay still either. I'm doing multiple versions, at least 2 versions, 2 generation versions, for each of these customers we have. And with each newer generation, they increase the consumption, the usage of the XPU. As they gain confidence, as the model improves, they deploy it even more. […]  And that's why I say we progressively gained share.”

Conclusion: 

Broadcom’s AI networking growth of 170% YoY and sharp rebound in custom accelerators at 63% YoY was certainly a highlight. In addition to strong AI growth, the company maintained a 67% EBITDA margin despite lower-margin hardware scaling, while record free cash flow of $7.0B highlighting the company’s financial strength. With AI now making up a third of total revenue and Q4 revenue projected at $6.2B, Broadcom is slated to accelerate XPUs as a primary beneficiary of the inference era.  

The nod toward accelerating growth in 2026 is why the stock went from flat to +5% AH, as the CEO seemed to hint that 70% AI growth or higher is not out of the question. Essentially, there are still four months to go in 2025 and Broadcom is gearing up for a strong 2026 already. The backlog of $110 billion was a “sit up in your seat” moment as it will force analysts to ponder – just how long will it take to work through that backlog? One can safely assume the backlog will only grow from here, as Broadcom is communicating they are preparing to be a strong contender to Nvidia toward the end of this decade.  

That’s a wrap! The I/O Fund’s earnings season is officially coming to a close. Keep an eye out for deep dives over the coming weeks plus coverage of notable earnings reports on stocks we don’t own.

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 AVGO at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• Credo’s Hypergrowth Intact as 274% YoY Growth Leads to Massive Bottom Line Expansion
• Dell Q2: Exceptional AI Growth yet AI Margins Miss the Mark
• Nvidia Q2: Guidance for Q3 Saved the Day; $10T Market Cap Prediction Revisited
• AMD Reports in Line while AI Story to Improve from Here

Nvidia is the global AI compute leader and there is not a distant second. Therefore, reporting (1%) for Q2’s compute segment should have tanked the stock. Instead, we are seeing a mild reaction because Q3 was quite strong and spells good things to come for Nvidia in the second half of the year.  

Given we are on the cusp of Blackwell starting to drive the stock’s narrative (c’mon already), it’s a good time to pause and talk about where Nvidia could go from here. I revisited the medium-term and longer-term forecast on a few pre-earnings discussions and was pleasantly surprised that Jensen Huang did the same on the earnings call.  

Also important for I/O Fund members who hold AI networking positions, the report was much stronger than expected with networking up a whopping 46% QoQ and 78% YoY to $7.25 billion. This foreshadows what’s to come in the AI networking space; and should be a boon for stocks like Astera Labs and Credo which supply Nvidia. 

Despite Q3 putting Nvidia right on track, I would not be surprised if the market offers soft price action on the stock in the near-term. The China narrative is confusing (and beaten to death), Q2 should mark a bottom and semiconductors don’t do well at their cyclical bottoms, plus the AI market is running on fumes until Nvidia can carry the market (Q3-Q4). 

First, we will run through the financials, and then in the second section I’ll break down what I think you can expect on this stock for the remaining part of this year, into calendar year 2026 and by the end of the decade. 

Slight Revenue Beat in Q2, Solid Q3 Guide 

Nvidia reported $46.74 billion in revenue in Q2, slightly ahead of estimates for $46.13 billion. This corresponded to growth of 55.6% YoY, decelerating more than 13 points from 69.2% in Q1; on a QoQ basis, revenue increased just 6.1%, slowing from 12% in Q1.  

This is expected to be a temporary lull in sequential growth, as Nvidia guided for $54 billion in revenue, +/- 2%, for Q3, corresponding to 53.8% YoY growth and a rebound to 15.5% QoQ growth. This was ahead of estimates for $52.7 billion, even with Nvidia noting that this guidance assumed no H20 sales in the upcoming quarter.  

For FY26, Nvidia is currently expected to report 55.4% revenue growth to $202.8 billion, though it is likely that this figure is revised higher, towards $205 billion, accounting for the slight beat in Q2 and Q3’s guide.  

Key Segments: First Sequential Decline in Compute Revenue, No China Sales 

Nvidia’s data center revenue increased 56% YoY and 5% QoQ to $41.1 billion, a marginal miss versus estimates for ~$41.2 billion in the quarter. This is also the smallest sequential increase since Hopper’s breakout quarter at just ~$2 billion.  

Notably, Q2’s data center result would mark the segment’s first miss since Q4 FY23 as Q1 was essentially in-line. 

While Compute revenue was up 50% YoY, the segment reported an unusual (1%) QoQ decline to $33.8 billion, a sharp shift from double-digit growth in late FY25 and 5% in Q1. Nvidia said the sequential decline was driven by a $4 billion QoQ reduction in H20 sales, as it reported no H20 sales to China this quarter and $0.65 billion in H20 sales to an unrestricted customer outside of China.  

On a more positive note, Nvidia said that it is continuing to ramp Blackwell and now Blackwell Ultra GPUs, with growth of 17% QoQ. Based on comments from Q1 placing Blackwell revenue in the $23.5-24B range, this would place Blackwell revenue approaching $28 billion.  

Networking growth was also robust in Q2, helping offset the softness in Compute as revenue rose 46% QoQ and 78% YoY to $7.25B. This marked a 22 point acceleration from 56% YoY growth in Q1. Nvidia said the strong performance in Networking was driven by “growth of NVLink compute fabric for GB200 and GB300 systems, the ramp of XDR InfiniBand products, and adoption of Ethernet for AI solutions.” 

• Gaming revenue was $4.29 billion, rising 14% QoQ and accelerating 7 points to 49% YoY. Nvidia said this was driven by sales of Blackwell-based GPUs. 
• Automotive revenue rose 69% YoY and 3% QoQ to $586 million. 
• Pro Viz revenue rose 32% YoY and 18% QoQ to $601 million. 
• OEM and other revenue rose 97% YoY and 56% QoQ to $173 million. 

Margins Outperform in Q2 with Strong Expansion 

Despite the softness in data center compute, Nvidia outperformed on margins, beating its guidance across the board and delivering a notable uplift in operating margin. Nvidia guided for more sequential improvement in operating margin in Q3, to the highest level since Q1 FY25. 

• GAAP gross margin was 72.4% in Q2, slightly ahead of guidance for 71.8%. Adjusted gross margin was 72.7%, or 72.3% excluding $180M in H20 inventory releases, ahead of guidance for 72%. 
• For Q3, Nvidia guided for GAAP gross margin to improve nearly 1 point to 73.3%, +/- 0.5%, and adjusted gross margin to improve to 73.5%, +/- 0.5%. 
• GAAP operating margin was 60.8% in Q2, improving nearly 12 points QoQ and coming in 1.7 points ahead of guidance for 59.1%. Adjusted operating margin was 64.5%, also up nearly 12 points QoQ and 1.4 points ahead of guidance for 63.1%. 
• For Q3, Nvidia guided for GAAP operating margin to expand 1.6 point QoQ to 62.4%, signaling continuing operating leverage tailwinds in the quarter. Adjusted operating margin was guided at 65.7%, up 1.2 points QoQ. 
• GAAP net margin was 56.6%, up 14 points QoQ and more than 6 points ahead of guidance for 50.2%. Adjusted net margin was 55.2%, up 10 points QoQ. 
• For Q3, Nvidia’s guidance implies a GAAP net margin of 52.9%, moderating 3.7 points QoQ. 

Slight Adjusted EPS Beat in Q2 

Nvidia reported just a 3.9% adjusted EPS beat in Q2, reporting $1.05 in earnings versus estimates for $1.01. This corresponds to growth of 54.4% YoY, rebounding substantially from Q1’s H20-affected 32.8% growth. Adjusted ESP growth is expected to remain strong through the rest of the fiscal year, at 47.2% and 53.2% in Q3 and Q4.  

For FY26, analysts are currently expecting Nvidia to earn $4.36 in adjusted EPS, up 45.8% YoY. This has improved $0.20 since May’s estimate of $4.16, or 39.2% YoY growth. While still early, Nvidia’s earnings growth is expected to remain quite strong in FY27 at 38% YoY to $6.02. 

Cash Flow Margins Drop to Lowest Level since Q4 FY23 

Nvidia’s cash flow margins dropped to the lowest level since Q4 FY23, as sharp QoQ growth in accounts receivable and inventories weighed on both operating and free cash flow. 

• Operating cash flow was $15.37 billion, down from $27.41 billion in Q1. OCF margin was 32.8%, down nearly 30 points QoQ and more than 15 points lower YoY.  
• Free cash flow was $13.45 billion, down from $26.14 billion in Q1. FCF margin was 28.8%, again down more than 30 points QoQ and more than 16 points lower YoY. 
• Accounts receivable surged $5.7 billion QoQ, or nearly 26%, to $27.8 billion. Nvidia said this was driven by timing of cash collections and Blackwell Ultra ramping late in the quarter. 
• Inventories rose more than $3.6 billion QoQ, or 32%, to $14.96 billion, to support Blackwell Ultra’s ramp. 
• Cash and equivalents totaled $56.8 billion, while debt remained steady at $8.47 billion.  

$200 Billion Run Rate (and why it matters) 

The Q3 guide was key as if we assume a similar mix of data center to other segments, Nvidia’s Q3 guidance puts it on the brink of achieving a $200B data center run rate with Q3 data center at $48.5B DC at midpoint.  

We had outlined more than a year ago in our free analysis that analyst estimates were too low, Nvidia Q1 Earnings Preview: Blackwell And The $200B Data Center stating: “These are the current estimates, yet if the analysts are correct, then the far right of the graph will end in $50B quarterly revenue. The difference between the current consensus and this much higher trajectory can be summarized in one word: Blackwell.” At the time, consensus was that we end the year with $33B in revenue. 

In fact, if we grow 10% QoQ in Q4 (it’s likely to be higher), then the data center will have quarterly revenue of $53.4 billion or about 62% higher than where analyst estimates were in May of 2024. This is a substantial disconnect for the world’s most valuable company (and a profitable one for investors who track this). 

The reason this matters is that if we nail the $50B quarter (very likely we do at $53.4B) then we are on track to hit a $75 billion data center quarter by the end of fiscal 2027. This assumes a 50% CAGR over 6 quarters and 10%-11% QOQ data center growth. Keep in mind, capex just grew 23% QoQ – which helps illustrate why 10% to 11% QoQ should be doable.  

If we do reach a $75 billion data center quarter compared to the $47 billion quarter that was recently reported, then that’s tracking about 60% growth in the AI segment. When you layer-in that we like to buy low when we can, and the market likes to sell semiconductors stocks at cyclical lows (when it’s really the best time to buy) we may see even more upside. 

Note: I elaborated on this pre-earnings in various interviews this week including Bloomberg, Schwab and Fox Business. Bloomberg, Schwab and Fox Business.  

$6 Trillion Market Cap Prediction – Revisited 

Nvidia has a 30 forward PS 3-year median but even if we go with something more reasonable like a 20 forward PS, then the market cap would be $6 trillion for $300 billion data center revenue.

From my perspective, it’d be better to get the stock lower if there is high confidence in reaching this market cap.

$10 Trillion Market Cap – Revisited 

To get to a $10 trillion market cap at a 20 forward PS, Nvidia’s revenue would have to be $500 billion. Analysts have the company reaching this a little after the year 2033 whereas I believe it will reach $10 trillion before 2030.  

The reality is that with Nvidia’s trajectory, we could see this market cap as early as 2028. That’s because we have not factored in software to the equation, which is expected to be an equal size market as hardware. The timing on software is tricky, and thus the I/O Fund has primarily preferred lower risk, yet also hypergrowth AI hardware stocks. 

Nvidia essentially has to grow the data center at a 30% CAGR for two years from 2026-2028 to reach my prediction of $10 trillion market cap two years early.  

That would mean analyst estimates are five years off as analysts have Nvidia reaching this revenue into the first quarter of 2034.

Note that I'm not stuck on this happening – we will shift if the market shifts. However, it leads back to why Q3 puts us right on track as the $50B data center quarter is achievable this year (the first milestone for these predictions). 

What Jensen Huang said in the Earnings Call 

I’d call this the total addressable market (TAM) earnings call as that was the predominant theme.  

In the opening remarks, JH stated: “We see $3 trillion to $4 trillion in AI infrastructure spend in the — by the end of the decade.” This was followed up by commentary that he sizes the market at $600 billion today. The math there is 5X growth.  

These are strong comments and thus it was followed up on during the Q&A where JH expanded his comments to say: 

“And so over the next couple of years, you're going to — well, you asked about longer term. Over the next 5 years, we're going to scale into with Blackwell, with Rubin and follow-ons to scale into effectively a $3 trillion to $4 trillion AI infrastructure opportunity. The last couple of years, you have seen that CapEx has grown in just the top 4 CSPs by — has doubled and grown to about $600 billion. So we're in the beginning of this build-out, and the AI technology advances has really enabled AI to be able to adopt and solve problems to many different industries.” 

This was followed up on by an analyst who asked for more specifics about what Nvidia’s share of that would be. Frankly, my guess would have been about 30% to 50% for Nvidia but management stated it would be closer to high 50% to 70%. 

“Benjamin Alexander Reitzes 

Jensen, I wanted to ask you about your $3 trillion to $4 trillion in data center infrastructure spend by the end of the decade. Previously, you talked about something in the $1 billion range, which I believe was just for compute by 2028. If you take past comments, $3 trillion to $4 trillion would imply maybe $2 billion plus in compute spend. And just wanted to know if that was right and that's what you're seeing by the end of the decade. And wondering what you think your share will be of that. Your share right now of total 

infrastructure compute-wise is very high, so I wanted to see. And also if there's any bottlenecks you're concerned about like power to get to the $3 trillion to $4 trillion. 

Jen-Hsun Huang 

[…] United States represents about 60% of the world's compute. And over time, you would think that artificial intelligence would reflect GDP scale and growth and so — and would be, of course, accelerating GDP growth. 

And so our contribution to that is a large part of the AI infrastructure. Out of a gigawatt AI factory, which can go anywhere from $50 billion to plus or minus 10%, let's say, $50 billion to $60 billion, we represent about $35 billion plus or minus of that and $35 billion out of $50 billion per gigawatt data center.” 

My thoughts are that this could be “CEO math” which is often too high. For example, JH had said 1K racks were shipping per week yet the reality is this print didn’t confirm that statement was true. The CFO did state that in this earnings call, and naturally, CFOs tend to be more accurate so let’s see if the 1K rack statement comes to fruition in the next earnings report. 

Regarding what was stated above, if the current number is $600B for AI infrastructure and we have close to a $200B run rate, then 33% is a good solid number to work with, especially given energy will take front and center on costs as we go along. At $3-$4 trillion, JH is basically stating Nvidia could have a $1 trillion data center segment if Nvidia sees 33% of that by the end of this decade. That would be up from a $184B data center segment today and up from a $200B data center segment in Q3. 

Additional Commentary: 

Management drew attention to the magnitude that demand outstrips supply: “Right now, the buzz is — I'm sure all of you know about the buzz out there. The buzz is everything sold out. H100 sold out. H200s are sold out. Large CSPs are coming out renting capacity from other CSPs. And so the AI-native start-ups are really scrambling to get capacity so that they could train their reasoning models. And so the demand is really, really high” 

In terms of why Rubin can stand its own against previous generations of GPUs and continue to drive forth demand, management expanded on the advent of reasoning agentic AI: 

“At the highest level of growth drivers would be the evolution, the introduction, if you will, of reasoning agentic AI. Where chatbots used to be one shot, you give it a prompt and it would generate the answer, now the AI does research. It thinks and does a plan, and it might use tools. And so it's called long thinking; and the longer it thinks, oftentimes, it produces better answers. 

And the amount of computation necessary for 1 shot versus reasoning agentic AI models could be 100x, 1,000x and potentially even more as the amount of research and basically reading and comprehension that it goes off to do.” 

Lastly, keep an eye on networking as it is a subtle clue that the NVL72s are starting to move the needle – above all else, we want to see these larger systems shipping as these massive SKUs are already at $28B this quarter and have a long runway to go. As you know, we are positioned to capture the AI networking tailwinds from these larger systems: 

“Networking delivered record revenue of $7.3 billion, and escalating demands of AI compute clusters necessitate high efficiency and low latency networking. This represents a 46% sequential and 98% year-on-year increase with strong demand across Spectrum- X Ethernet, InfiniBand and NVLink. Our Spectrum-X enhanced Ethernet solutions provide the highest throughput and lowest latency network for Ethernet AI workloads. Spectrum-X Ethernet delivered double-digit sequential and year-over-year growth with annualized revenue exceeding $10 billion. At Hot Chips, we introduced Spectrum-XGS Ethernet, a technology design to unify disparate data centers into giga-scale AI super factories. [ CoreWeave ] is an initial adopter of the solution, which is projected to double GPU-to-GPU communication speed.” 

Conclusion: 

I wouldn’t be surprised if the market continues to sell this report, as the market likes to do at cyclical lows for semiconductor companies. What’s quite impressive is that Nvidia has continued to beat even at its cyclical low (defined as being in-between GPU generations) and despite the curveball with China’s H20 revenue. 

Our fundamentals process favors QoQ acceleration and the 15% QoQ growth for Q3 helps put our estimates right on track. The strong networking growth spells good things to come as it indicates it’s the larger systems that are propping up the growth (Blackwell was stated to have grown 17% QoQ).  

Now is the perfect time to pause and look a the medium-term and longer-term picture to answer for ourselves if the world’s most valuable company can continue to grow – or are we in a bubble? Each Member will need to decide that for themselves although my answer is “yes, for many AI stocks we are in a bubble – but for Nvidia, we are not.” Hopefully, this post-earnings writeup helps to substantiate why I continue to believe there is room in this stock.

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:

• Coherent Q4: Data Center Growth Slowing QoQ; Competitive Concerns
• Cloudflare Q2: Multi-faceted AI Positioning, Steady Growth
• AMD Reports in Line while AI Story to Improve from Here
• Astera Labs Q2: Blowout with double digit Beat/Raise; Emphasis on future growth

Astera Labs reported a blowout quarter with beat/raise that is impressive even for ALAB’s standards. Not only did Astera Labs beat by nearly $20M or 11.3% but the guide exceeded this and beat by 14.35%. 

To sweeten the deal, the roughly $20M beat flowed through to profits for a company that is comfortably GAAP profitable despite a fairly recent IPO in 2024 with elevated stock-based compensation at 20% of revenue.  

The fundamental profile of Astera Labs could not be cleaner, and when it comes to its products, the company said all the magic words such as ramping in volume, new design wins with ten customers including merchant GPUs and ASICs, plus a line of sight to further sales growth from their many product lines in H2 2025 and 2026, and the upcoming UALink consortium in 2027. 

To illustrate, in the opening remarks, the company pointed toward the drive for low latency PCIe as the primary contributor to the beat/raise across both the Aries and Scorpio products. Launched only this year, Scorpio now exceeds 10% of total revenue “making it the fastest-ramping product line in Astera Labs’ history.”  

Management also stated there were new design wins across multiple customers this quarter with PCIe6 solutions beginning “volume ramp during the quarter within rack-scale merchant GPU-based systems.” This is music to our ears as it indicates ALAB is a direct beneficiary to the highly anticipated Blackwell systems. More specifically, it was stated the Scorpio smart fabric switches transitioned to volume production in Q2 for PCIe6 scale-up applications “deployed within GPU customized rack scale solutions.” 

UALink was also discussed at length, which is the open source alternative to proprietary interconnect protocols such as Nvidia’s NVLink. The data rate specifications of up to 200G per lane combined with the low latency of PCIe6 puts Astera in an “excellent position.” Per management: “As a leading promoter of UA Link, Astera Labs is committed to developing and commercializing a broad portfolio of UA linked connectivity solutions ranging from AI fabrics to signal conditioning solutions and other I/O components. Proliferation of UA Link in 2027 and beyond will represent a long-term growth vector for Astera Labs.” 

In terms of quantifying where this could lead, management stated: “scale up connectivity for rack-scale AI infrastructure alone will add close to $5 billion of market opportunity for us by 2030” although truly the opportunity could be higher given it was also stated: “our silicon dollar content opportunity has expanded into the range of multiple hundreds of dollars per AI accelerator which has effectively established a new revenue baseline for the company.” From there, it was further stated that Scorpio-X will cause the dollar content opportunity to increase even further: “Given the extreme importance of scale-up connectivity to overall AI infrastructure performance and productivity, we see Scorpio X Series solutions as the anchor socket with the next-generation AI ranks. We are engaged with over 10 unique AI platform and cloud infrastructure providers who are looking to utilize our fabric solutions for their scale-up networking requirements. We look for Scorpio series to begin shipping for customized scale-up architectures in late 2025, with a shift to high-volume production over the course of 2026.  

With the ramp of Scorpio X Series for scale-up connectivity topologies next year, we expect our overall silicon dollar content opportunity per AI accelerator to significantly increase. Overall, we expect this to be another step-up from a baseline revenue standpoint. Also, given the sale of the scale up connectivity opportunity, we expect our Scorpio X Series revenue to quickly outgrow Scorpio P-Series revenue. In 2026 and beyond, cloud platform providers and hyperscalers will begin to deploy next-generation platforms as the industry transitions to AI infrastructure 2.0.” 

Usually, I put these bigger quotes at the bottom under the Q&A yet I didn’t want our Members to miss the emphasis the earnings call had on future growth. But be sure to not miss the additional strong commentary regarding H2 2025, 2026 and 2027 noted below.  

We’ve covered more on Astera’s products and positioning in this analysis here and here.analysis here and here. 

Revenue Beats by 25 Points, Guide Beats by 22.8 Points (wow!) 

Astera Labs reported revenue of $191.9 million, beating consensus of $172.5 million for growth of 150% YoY and 20% QoQ. About eight months ago in November, analyst consensus for the June quarter was for 85% growth — thus the company has nearly doubled these expectations in less than a year. Going into the print, analysts had been steadily raising consensus for growth expectations of 124.4% and Astera beat these estimates by 25 points. This technically marks accelerating growth from last quarter, which reported growth of 144.3%. 

For the September quarter, the beat is also pronounced on a YoY basis although less so on a QoQ basis given the strong beat this quarter. Management guided for revenue of $206.5 million at the midpoint for growth of 82.6% YoY and 7.6% QoQ. Going into this print, analysts were expecting 59.8% growth, with Astera beating by 22.8 points.  

For perspective, about eight months ago, Astera was expected to report $157.5 million for 39.3% growth in the upcoming September quarter. Today, Astera is guiding for more than double growth expectations from less than a year ago. I’m presenting these historical estimates to help illustrate just how quickly Astera has skyrocketed from ideal positioning in the AI networking stack. We’ve covered more on Astera’s products and positioning in this analysis here and here. 

Margins and EPS: The $20M beat flows directly to profits 

The only thing that can sweeten the deal of a 25-point top line beat is a similarly strong bottom line beat. Astera delivered in that regard with a GAAP operating margin of 20.7% compared to 7.9% expected. This led to GAAP profits of $39.8M compared to $13.7M expected, helping to see the operating leverage as the $19.4M beat flowed nicely down the income statement (plus some).

Here’s a visual on how the GAAP profile of ALAB has rapidly improved: 

• Gross margin of 75.8% was up 90 bps from last quarter yet was down 210 bps from the year ago quarter. This led to gross profits of $145.6M, beating the guide for $127.7M. Adjusted gross margin was similar at 76%. 
• GAAP operating margin of 20.7% is a major win for ALAB investors as the company is now comfortably GAAP profitable despite stock based compensation being around 20% of revenue. 
• Adjusted operating margin of 39.2% compares to the guide for 31.1% leading to adjusted operating profits of $75.3 million. This is significantly higher than the adjusted operating profits of $18.7 million in the year ago quarter.  
• The GAAP net margin of 26.7% for net income of $51.2M is significantly higher than the (9.8%) net margin from a year ago and represents a 670 bps improvement QoQ.  
• The adjusted net margin of 40.7% is up from 28.9% last year and 330 bps improvement QoQ.

Cash: 

Astera’s cash from operations increased significantly with an operating cash flow margin of 70.5% up from 38.7% last year. This totaled operating cash of $135.4M with $1.07B in cash on the balance sheet and no debt. 

The company provided a 6 month statement for its cash flows rather than a quarterly statement, but it’s easy enough to deduce that the majority of operating cash flow went to the cash reserves as free cash flow. On a six month basis, the company had free cash flow of $139M with $6M last quarter, leaving about $133M in free cash flow this quarter for a FCF margin of 69.3%.

Earnings Q&A: 

More commentary regarding future growth: 

To say you have 10 customers (who we can infer are at massive scale) is a strong statement for a company like Astera Labs sitting at a run rate of roughly $800M. Buried in the call, management clarified these customers are “nearer-term opportunities that we are tracking based on PCIe.” 

What Astera is referring to is that PCIe 6.0 doubles the bandwidth of PCIe 5.0, reaching up to 256 GB/s. This is crucial for Blackwell GPUs, as the increased bandwidth translates into faster data transfers for quicker processing and reduced latency when training an AI model training and for inference. 

At the end of the call, it was also clarified the way these customers will ramp should help growth into the foreseeable future: 

Sujeeva De Silva   ROTH Capital Partners 

Helpful. And then my follow-up on Scorpio X, you talked about 10 customer engagements. I'm wondering if that implies multiple programs per customer, if they're going to think about using you standard in their platforms? Any color on how those are kind of shaping up would be helpful in programs versus customers. 

Sanjay Gajendra   Co-Founder, President, COO & Director 

Yes. So 10 plus we noted are unique customers now within each customer, there are multiple opportunities that we're tracking. Some of them are design wins, and some of them are ramping to production. Some of them are design ins going through qualification. Some of those are early engagement. So in general, we are very pleased with the amount of traction that we're seeing for our Scorpio family.”

There was additional commentary offered in terms of how they plan to retain these customers and perhaps even increase their sales during the transition to UALink open standards: 

“And we do have, like I noted, several customers, we are counting 10 plus right now that are looking at leveraging some of these open standards, whether it's PCIe in the short term, combination of PCIe and UA Link in the midterm and transitioning perhaps to a broader UA Link deployment in 2027 and later. So overall, I think the momentum is shifting positively, and we are excited to be in the middle of it and driving the adoption of open and scalable supply chain in the market.” 

Later, it was clarified the ten customers are primarily for Scorpio X-Series – which has not even ramped yet – spelling good things for future growth. 

“This, we see, like you noted, as an anchor socket because that is truly the socket that holds all the GPUs together, and today, like we noted, we have 10 plus customers that we are engaging when it comes to scale up networking using Scorpio X-Series.”  

It was later stated the X-Series will grow beyond Scorpio P-Series which is what is driving the growth right now: “For the X Series, we do have preproduction volumes here, but really, that starts to go into high volume production during the course of 2026 and layering even more growth. Ultimately, what we called out is the X Series is going to grow to be bigger than P-Series" 

There were a few additional comments pointing toward strong growth in 2026: 

“The scale up this year is predominantly preproduction volumes. And these systems are pretty complex that they're shipping into. So we like to try to be conservative on how we telegraph those going forward. But the volume opportunities scale up connectivity for switching is a much bigger dollar opportunity for us as we look forward. But those designs really will start to enter into full volume production during the course of 2026. So not a driver in the next couple of quarters.” 

Conclusion: 

Astera is a strong contender for best earnings report of the quarter – not only in I/O Fund’s portfolio but broadly speaking across the tech sector (the other contender across the tech sector is Reddit).  

This company is firing on all cylinders with unique positioning in the AI economy as they serve both merchant GPUs and custom silicon – which is what makes 10 customers a possibility.  

What I liked most about this call was management clearly outlining how they plan to drive future growth across its product lines, plus across a diverse set of customers throughout 2026 – and they furthered the discussion on how ALAB will strategically continue to drive growth during the anticipated transition to UALink in 2027.  

The note from our last portfolio meeting following my Top 15 Stocks report was “is 20% allocation too high for Astera Labs?” Apparently not. It’s at 16% now, and by default, will open at a higher allocation tomorrow.

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. Beth Kindig and the I/O Fund own shares in “ALAB” at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• The I/O Fund’s Top 15 Stocks for Q3 2025
• Coinbase Underperforms in Q2; Near-term Volatility Likely
• Meta Reports Large Q2 Beat, Ad Impressions Growth Rebounds
• Microsoft FYQ4: One of the Strongest Earnings Reports in Multi-Decade History

AMD reported in line, yet the market’s short attention span has likely forgotten the QoQ decline in the data center was expected. Per our last earnings report writeup: 

“For Q2, data center will decline due to the MI308 revenue being excluded. When asked about future quarters, the CEO Lisa Su stated the DC segment would resume growth after Q2: “in Q2, it's not going to grow year-over-year just given what we've said about the $700 million coming out of Q2 and how we had previously talked about the evolution. But we do believe that we'll grow year-over-year going forward, in Q3 and Q4 certainly, for us to do the full year with strong double-digit growth.” 

Therefore, what matters for the purpose of our position is if AMD can execute and drive much better data center segment results in Q3 and Q4. For example, in the opening remarks, it was stated while citing the Oracle deal which includes 130,000 MI355s: 

“We began volume production of the MI350 series ahead of schedule in June and expect a steep production ramp in the second half of the year to support large-scale production deployments with multiple customers.” 

We go through the obligatory financials below while noting from the earnings call additional comments about the one thing that really matters – that AMD executes in H2 and further executes in 2026.  

Revenue beat driven by Client and Gaming segments: 

AMD reported a slight beat on the top line at $7.685B in revenue compared to estimates of $7.43B. This represents growth of 31.6% compared to growth of 27.4% expected. Where the beat is somewhat problematic is that it was driven by Client and Gaming, rather than the data center. 

However, in terms of a bright spot in the report and a data center inflection point, Q3 estimates were at $8.3B going into the print and management is guiding for $8.7B at the midpoint for growth of 28% YoY and 13% growth QoQ. This will be driven by data center, as was alluded to on the earnings call: “Sequentially, we expect revenue to grow by approximately 13%, driven by strong double-digit growth in the Data Center segment with the ramp of our AMD Instinct MI350 series GPU products […]” with Client expecting to see only modest growth. 

In addition, AMD estimates have been steadily rising after a trough of sorts earlier this year. For example, the September quarter was expected to see as low as 17% growth per consensus in May, yet is now at 28% per management guidance.  

Revenue Segments: Data Center declines QoQ from China Loss 

Last quarter, management had stated, “we expect data center segment to decrease due to the exclusion of MI308 revenue.” Therefore, it was not a surprise when data center was down (11.8%) QoQ yet was up 14% YoY for revenue of $3.24B. This compares to DC revenue of $3.67B last quarter and $2.84B last year.  

Here is what it looks like on a YoY basis: 

Gaming and Client exceptionally strong in Q2 

While many consumer device companies are struggling right now, AMD is breezing past consumer demand concerns with incredibly strong Client and Gaming revenue. Whether this can sustain or if it was a pull forward remains to be seen, with management guiding for Q4 to be seasonally weaker than usual. 

• Client revenue of $2.5B up 9% QoQ and up 68% YoY 
• Gaming revenue of $1.1B up 73.4% QoQ and up 73% YoY 
• Embedded revenue of $824M, flat QoQ and down 4.5% YoY 

According to the opening remarks, it was not a pull forward rather the popularity of its Ryzen processors and Radeon 9000 GPUs that drove the strong performance.  

Regarding Client CPUs, it was stated: 

“We delivered record desktop channel CPU sales as Ryzen processors consistently topped the best-selling CPU lists at major global e-tailers throughout the quarter [..] In mobile, demand for AMD-powered notebooks was strong with sellout growing by a large double-digit percentage year-over-year. We drove a richer mix of higher ASP mobile parts year-over-year as we expanded our share in the premium notebook segment where our Ryzen AI 300 CPUs deliver leadership performance and value for both general purpose and AI workloads. In commercial PCs, Ryzen adoption accelerated as OEM consumption increased more than 25% year-over-year.” 

Regarding the Radeon series which drove 74% QoQ growth in Gaming, there were partnerships with Microsoft/Xbox and Sony. The following was also stated: “In PC gaming, demand for our latest-generation Radeon 9000 series GPUs was very strong, with desktop GPU sell-through accelerating in the quarter as demand outpaced supply.” 

Despite Client being strong this quarter, management cautioned this is inventory building for the holiday season and this segment will be down in the fourth quarter “strong double digits.” 

Margins down due to China; Expected to rebound quickly  

EPS was in line with expectations at $0.48 yet was down (30%) from $0.69 in the year ago quarter. The company is expected to rebound quickly with EPS of $1.15 next quarter.  

Gross margin of 40% is significantly lower than previous quarters in the 50% range. This represents profit of $3.1B due to $800M in inventory changes from expert controls. Management pointed out that minus the $800M, gross margin would have been 54%: “Excluding the $800 million inventory write-down related to data center AI export controls, gross margin was 54%, marking our sixth consecutive quarter of year-over-year margin expansion led by a richer product mix.” 

 Adjusted gross margin of 43% represents adjusted gross profit of $3.33B. Notably, the margins were guided correctly and in line with expectations following the loss of China revenue discussed in the previous quarter.  

Operating margin of (2%) for operating profits of ($134M) also included the $800M in inventory changes. Adjusted operating margin of 12% was guided correctly and was in line with expectations. Adjusted operating profits of $897M beat expectations for $882M.  

Net margin of 11.3% was 600 bps higher than the previous year and 230 bps higher than last quarter. However, adjusted net margin was down significantly by 10 points to 10.2%.  

Cash Flow margins Improve QoQ, Debt profile improves 

AMD’s cash flow margins sustained well at 20% operating cash flow margin compared to 13% last quarter and 10% OCF margin last year. Free cash flow margin of 15% also expanded from a year ago at 8% margin and up from 10% FCF margin last quarter.  

AMD has cash of $5.9B on the balance sheet and debt of $3.2B, down from $4.2B last quarter.

Earnings Call Q&A: 

Key points on how AMD plans to execute from here: 

Sovereign AI to pick up in 2026: 

In the opening remarks, management discussed its “multibillion-dollar collaboration with HUMAIN to build AI infrastructure powered entirely on AMD CPUs, GPUs and software.” The HUMAIN deal refers to a $10 billion joint venture between Saudi Arabia and AMD to create an AI hyperscaler for the country’s sovereign AI initiatives.  

According to Lisa Su during the Q&A: “So look, we're really excited about the overall AI opportunity for us with MI355 and the MI400 series as we go through the back half of this year and into 2026 […] I think you heard from Tareq that was — he was at our event saying that, that would start with MI355, that we would expect that, that would continue on. I think what's attractive about our offering is our open ecosystem, and I think that really resonates with the sovereign community. But to your original question, I think it's an additive opportunity, and it's one that we believe will continue to be very important for us going forward with both MI355 as well as the MI400 series.” 

MI350s starting to ramp: 

When asked about the size of opportunity from the MI300s and MI350 Series this year (mainly the MI350s) and if this can get to $7B, management declined to be specific yet stated: “I think what we said in the prepared remarks is that we are seeing a strong ramp from Q2 into Q3. MI355, we actually started production in June. So we had some shipments sort of in the month of June, but it really is ramping as we go through this quarter and the third quarter. So in terms of guideposts, we said it would grow year-on-year from last year. And that, I think, is a strong ramp, and then we would expect it to grow into the fourth quarter as well.”

MI400s to ramp next year including Helios: 

The MI400 series will be the start of rack-scale systems for AMD, starting with Helios, which will connect up to 72 GPUs similar to Nvidia’s NVL72 systems. According to AMD, Helios will “deliver up to a 10x generational performance increase for the most advanced Frontier models, and we believe it will be the highest-performance AI system in the world when it launches.” The last part is doubtful yet the effort to close the gap with Nvidia will likely go a long way when coupled with lower pricing.  

AMD’s goal of reaching tens of billions in MI400 sales was also elaborated on: 

“Joseph Moore   Morgan Stanley: 

You used this language before, the kind of tens of billions opportunity around MI400. Can you talk about the time frame when that might occur and not to pin you down too much, but — and what would help you get to that level sooner rather than later? Should we think of that as a 2027 realistic outcome that you could be looking at $20 billion-plus? Just a little bit more color on that tens of billions comment. 

Lisa Su   Chair, President & CEO: 

Yes. I mean, maybe without being specific, Joe, I can give you sort of the way I look at it and back to this notion of are we incrementally more confident. I think we're seeing a lot of positive signs in our AI customer adoption, I think the strength of the MI350 series, the very positive feedback that we're getting on MI400 from customers, the work that we're doing in terms of ensuring that we are fully ready for large-scale deployments of not just inference but training. 

I think when we get to tens of billions of dollars, we're talking about significant gigawatt-scale type deployments. And those would be important for us to get there. And we're certainly, I think, engaged with all of the right customers that can enable that type of ramp. But I won't necessarily speculate on the exact time other than to say, certainly, that would be our set of aspirations.” 

It was later stated: “We would expect significant revenue contribution from Helios in 2026.” 

Conclusion: 

In my Top 15 stocks report the conclusion was the following: The risk to AMD is primarily in Q2’s data center growth decline, and how quickly can the company ramp its MI355s and subsequent MI400s while in the midst of Nvidia’s large shadow – will we see a solid surprise arrive in Q3, Q4 or even into next year? My best guess is the most meaningful AMD moment is not likely to occur during Blackwell’s NVL72s release – I think 2025 belongs to Nvidia and somewhere between 2026-2027 we switch it up. 

AMD has required some patience, but 40% returns YTD are not bad. We weren’t expecting much from this report given the concise management commentary from last quarter. However, we do foresee watching this stock very closely come 2026 with a placeholder in the portfolio should AMD surprise before then.

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. Beth Kindig and the I/O Fund own shares in “AMD” at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• Astera Labs Q2: Blowout with double digit Beat/Raise; Emphasis on future growth
• Microsoft FYQ4: One of the Strongest Earnings Reports in Multi-Decade History
• Taiwan Semiconductor Q2 Earnings: FY25 Guidance Raised on Strong AI Demand
• Can AMD’s MI350X and MI355X GPUs Close the Gap with Nvidia?

This article is a continuation of our free newsletter from June 19, AMD vs Nvidia: The AI Stock That Could Win by 2028

Find out the following below: 

• We compare AMD’s MI350X and MI355X with Nvidia’s B200s and GB200s to decipher if AMD has what it takes to close the gap with the AI leader 
• Clear conclusions on the next 1-2 years that are tailored for stock investors and how we plan to position our portfolio  
• The SKU that all investors should know about 

Last week, AMD introduced its Instinct MI350 series GPUs, including MI355X with up to 4X performance over the previous MI300X generation and up to 40% more tokens per dollar compared to Nvidia’s B200 accelerators. The company also previewed its Helios rack-scale server architecture featuring the MI400s for 2026 deployments. 

According to Tom's Hardware AMD is claiming the eight-GPU MI355X system is 1.3X faster than Nvidia’s DGX GB200s systems with Llama 3.1 and up to 1.2X faster than the B200 HGX systems in inference for DeepSeek R1 with equivalent performance as Llama 3.1 when tested at FP4. 

Here are a couple of key points in terms of how AMD is starting to close the gap with Nvidia for inference purposes: 

Floating point precision: 

AI accelerators are increasingly offering lower floating-point formats to help reduce memory consumption and bandwidth requirements, which in turn speeds up computation and lowers power consumption. For example, FP8 delivers better throughput and energy efficiency in LLM inference compared to FP16. The newer generations of GPUs will offer FP4 formats to further alleviate memory-bandwidth bottlenecks and improve performance for large matrix operations. 

I elaborated on the importance of floating-point precision in my analysis “Here’s Why Nvidia Will Reach a $10 Trillion Market Cap” when I stated: “The difference is that the smaller bit size allows for an economical way to achieve more speed when giving up a small amount of accuracy doesn’t make a critical difference. As discussed, this also helps in the face of a slowing Moore’s Law.” 

With the MI350X and MI355X, AMD is introducing FP4 along with the smaller formats of FP8, FP6 and FP4, which are especially helpful for inference. In the CDNA 4 architecture, the FP6 data rate shares the same peak PFLOP/s as FP4 — which for inference purposes means it will be comparable to or slightly exceed Nvidia’s B200s.  

ServetheHome states, “AMD is doing the higher performance (at a transistor cost) option of adding FP6 to the FP4 pipeline to give it a big boost.” 

Source: Tom’s Hardware, pictured above – FP6 performance is on par with FP4 performance.  

HBM3E and HBM4 Memory: 

AMD is attempting to compete on memory by slightly beating Nvidia with the MI355X having 1.6x more memory capacity than the B200s. This allows AMD to load full model weights into memory for fast inference and avoids having to share resources between multiple GPUs. The higher amount of memory also increases the batch size, which increases throughput while lowering latency.  

It’s important to keep in mind that Nvidia is preparing to send a shockwave through the AI market, once again, with its NVL72 and NVL36 systems. These systems combine 72 GPUs and 36 GPUs to think like one GPU, which I’ve covered recently here.covered recently here. 

Rather than AMD taking head-on Nvidia’s NVL72s and NVL36s right now – which are earth-shattering SKUs — the company is instead attempting to compete at the 8-GPU system level. Memory is a big part of that attempt. Inference craves low latency, thus having the model fit entirely in memory for inference purposes is a part of that strategy.  

What’s Important About the MI350X and MI355X: 

To put it plainly, on the AI accelerator front, this will be the first time that AMD will overlap Nvidia in terms of benchmarks on GPUs. Please do note, the amount of time that AMD’s current generation of GPUs and Nvidia’s GPUs overlap will be brief – and will only be at the single GPU and 8-GPU system level. AMD was originally expected to ship the MI350s at the end of this year yet are moving the shipments up – which fits with AMD’s tradition of underpromising and overdelivering.  

However, the accomplishment is noteworthy as it’s setting the tone as the inference market begins to ramp. In other words, AMD ceded the training market to Nvidia – but I do not expect that to be the case with the inference market. 

When Blackwell Ultra ships, the B300s will offer FP4 TFLOP/s that is 1.3X faster than AMD’s current MI350X and MI355X. With that said, because AMD has prioritized competing on memory — its bandwidth and capacity is expected to be on par with Blackwell Ultra. 

AMD’s CDNA 4 Architecture: 

The primary architectural changes of CDNA 4 were aimed at increasing memory capacity and bandwidth per compute unit. The lower precision compute capacity was also increased, favoring FP6 and FP4. 

AMD’s architecture is built on a chiplet design, and similar to the Zen-2 architecture discussed above, the chiplet design offers power efficiency improvements from monolithic designs by offering a dozen chiplets on a single processor.  

Although monolithic used to be preferable, to compare, Nvidia’s has evolved its architecture to utilize multi-die modules (MCM) which combines two reticle-limit dies. By utilizing high bandwidth connections, the two dies function as a single die to forego reticle-size constraints, helping to improve yields and results in higher performance. 

However, keep in mind that AMD was first to market with chiplets in the Zen architecture that helped stage the company’s comeback. Nvidia is the world’s best AI semiconductor design company, yet the point is that AMD is not necessarily a follower. In some design areas, AMD leads. 

A few more things to highlight from last week’s announcements: 

• 3D packaging with CoWoS-S from the MI300s remains with XCDs, HBM3 memory, I/O Dies and the Infinity cache 
• There are a total of 256 compute units with eight 32 CDNA per XCD. This is less than the last generation yet with the 3nm, each compute unit delivers more power 
• There are two larger I/O Dies rather than four for better efficiency. The I/O Dies are built on a 6nm process. 
• More memory at 288GB of HBM3E with 8TB/s 

MI400s “Helios” Will Close the Gap on Larger AI Clusters 

The market is forward-looking, which means investors should be too. AMD is closing the gap on single GPUs and 8-GPU systems, yet the MI400s will mark a pivotal moment as AMD will attempt to compete on rack-scale systems with Helios, its 72-GPU systems. If things go as planned, AMD will be competitive with Nvidia on GPU, memory and interconnect performance — while potentially taking the lead on memory capacity and bandwidth.  

By using UALink and potentially Broadcom’s scale-up ethernet, AMD will be able to deliver considerable bandwidth, with projections of 31 TB of HBM4 memory and 1.4PB/sec of bandwidth, which would beat Nvidia’s offerings by 50%. 

UALink, or Ultra Accelerator Link, is an open industry-standard interconnect that enables high-speed and low-latency communication for AI clusters. This is a joint venture between a consortium of Nvidia competitors, including AMD, Intel and Broadcom, to take-on Nvidia’s proprietary NVLink. The first generation of UALink supports 1.28 TB/s of bandwidth for systems of 4 to 8 accelerators while future generations will support racks of 72 accelerators and more. 

Conclusion: 

Judging by the poor stock performance over the past 1-2 years, the market thinks AMD is down for the count. I think it’s the nuances of AI training versus inference (and timing of those markets) that has made AMD appear to be inconsequential to AI hardware. Although I do not foresee AMD surpassing Nvidia in terms of market cap by a long shot, I believe it’s highly probable that AMD’s returns outpace the AI leader due to the sheer amount of revenue growth hidden within AI inference. Specifically, inference is expected to be a larger market than training, and AMD’s strengths will finally be on full display.  

My current prediction is that AMD does not need to even come close to overtaking Nvidia on revenue or market cap for the stock performance to exceed Nvidia's over the next few years. Rather, the unforeseen second wind from GPUs and AI systems will be enough to make second place the most rewarding era in AMD’s history.

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 and AMD at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• This AI Stock is Set to Surge from Inference Demand — Broadcom
• Palantir Stock: Strong Sequential Growth and Strong Underlying Key Metrics
• Taiwan Semiconductor: Building a Moat under Geopolitical Tensions
• Dell Riding Nvidia’s Tailwinds to Record $12.1B in AI Server Orders in Q1
• Nvidia Q1/Q2 Guide: Blackwell is (Finally) Here