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

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

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

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

Astera Labs: Aries PCIe Gen 6 Tone Changed

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

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

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

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

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

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

The following was discussed in the Q&A section:

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

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

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

Thomas O'Malley   Barclays Bank

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

Michael Tate   CFO

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

Coherent: Tone on Pluggable Optics Versus Active Cable

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

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

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

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

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

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

Read-through for Credo?

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

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

Power Management Integrated Circuits (PMICs):

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

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

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

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

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

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

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

Conclusion:

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

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

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

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

Recommended Reading:

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

Astera Labs reported solid earnings that beat the top-line and bottom-line estimates driven by AI strength. Q4 revenue grew by 179.4% YoY and 24.8% QoQ to $141.3 million, beating estimates by 10.3%. Adjusted EPS grew by 208.3% YoY to $0.37 and 150% YoY to $0.30 after excluding one-time tax benefit, beating estimates by 15.8%.

Management Q1 revenue guide is $151 million to $155 million, representing YoY growth of 134.4% at the midpoint, beating estimates by 14.2%. Adjusted EPS guide is $0.28 to $0.29, beating estimates by 14%.

The only downside in the report was the gross margin, which was below the management guide due to the higher mix of hardware revenue. Management reiterated that margins will trend closer to the long-term gross margin target of 70% from the current 74%.

Looking ahead, the IOF is likely to build a bigger Astera position in 2025 due to nearly-unrivaled pricing power from PCIe 5.0 retimers and higher average sales prices from the soon-ramping Scorpio PCIe 6.0 switches. In our original analysis, it was stated: “We are interested in Astera Labs for the increased average sales prices that are expected to persist at least through 2025-2026 due to the Aries products and upcoming Scorpio products.” This was reiterated on the earnings call: “we will continue to benefit from increased dollar content per accelerator in these next-generation AI infrastructure systems.”

The see-saw market reaction is similar to what we are seeing across many AI-related hardware companies, which is due to mixed messaging around the delivery of Blackwell systems. Management indicated their Q1 is primarily being driven by custom silicon programs, which is a positive to be diversified, yet requires additional due diligence for Nvidia investors such as ourselves.

Overall, management teams are offering a more muted Q1 in their commentary with increased anticipation for H2 2025. Where Astera Labs is unique is they serve both custom silicon and GPU systems, of which, there are many systems combining both.

Regarding tariffs and China, Astera Labs has 15% exposure and this is less of a concern for this stock, then say, something like Blackwell being delayed. It’s speculative as to the extent that Blackwell is delayed, yet we have now another management team aligned with this possibility.

I look at this and more below.

Q4 Revenue Grows 179% from Aries and Taurus Products

The company’s Q4 revenue grew by 179.4% YoY and 24.8% QoQ to $141.3 million, beating estimates by 10.3%, driven by strong AI demand.

The company experienced significant revenue growth in Q4 primarily due to strong sales of its Aries PCIe Retimer and Taurus Ethernet Smart Cable Module products. Leo and Scorpio's product momentum continued, with both products shipping in pre-production volumes during the fourth quarter.

Here is what management stated about the beat – most intriguing is the note that 2025 will be a “breakout year” and that the Scorpio products are slated to become the company’s largest product line (and yet did not drive the current outsized revenue growth and beat):

“Our revenue growth in 2024 was largely driven by Aries products, along with a strong ramp of Taurus in the fourth quarter. We expect 2025 to be a breakout year as we enter a new phase of growth driven by production revenue from all four of our product families to support a diverse set of customers and platforms. We expect 2025 to be a breakout year as we enter a new phase of growth driven by production revenue from all four of our product families to support a diverse set of customers and platforms. 

In 2025, our Aries and Taurus retirements are on track to continue their strong growth trajectory. Also, Astera Labs is poised to be a key enabler of CXL proliferation over the next several years with the volume ramp of our Leo family expected to start in second half of ‘25. 

Finally, our Scorpio smart fabric switches will begin ramping this year with new and broadening engagements for a scale up with our X series and scale out with our PCD switches. In time, we expect Scorpio Fabrics switches to become our largest product line, given the size and growth of the market opportunity for AI Fabrics.”to become our largest product line, given the size and growth of the market opportunity for AI Fabrics.”

• Management Q1 revenue guide is $151 million to $155 million, representing a YoY growth of 134.4% and 8.4% QoQ at the midpoint, beating estimates by 14.2%.
• Analysts expect revenue to grow 108.6% YoY and 5% QoQ to $160.58 million in Q2 and 59% YoY and 12% QoQ to $179.85 million in Q3.

• 2024 revenue grew by 242.3% YoY to $396.3 million, accelerating from 45% growth in 2023.
• Looking further out, analysts expect revenue to grow 75% in 2025 and 25.8% in 2026.

Regarding 2025 being a breakout year, management stated it was partly due to increased dollar content per accelerator. We offer more in the Q&A section below, yet this was stated in the opening remarks: “As we look into 2025, we see strong secular trends across the industry, supported by higher CapEx spent by our customers, broadening deployment of AI infrastructure driven by more efficient AI models, and company-specific catalysts that should drive above market growth rates for Astera Labs.

Specifically, for 2025, we expect three key business drivers. One is the continued deployment of internally developed AI accelerator platforms that incorporate multiple Astera Labs product families, including Aries, Taurus and Scorpio. As a result, we will continue to benefit from increased dollar content per accelerator in these next-generation AI infrastructure systems.”

The total addressable market of $12 billion was further reiterated by saying, “We estimate our portfolio of hardware and software solutions across retimers, controller and fabric switches will address a $12 billion market by 2028.”

Gross Margin Contracts 330 Basis Points YoY

The company’s bottom line is growing at a remarkable pace. However, the margins will trend lower in the coming quarters due to the expected higher mix of hardware revenue. Notably, stock-based compensation is high yet the company has strong GAAP profitability potential as SBC becomes a lower percentage in the coming quarters/years.

The CFO answered an analyst question on margins for 2025. “Now, if you go into 2025, we still see good contribution from Taurus and Aries SCM modules. But as we make it through the year, the Aries board and chips as well as Leo and Scorpio are a positive for us as well. So, we think Q1 and Q2, we should have a consistent margin profile of around 74%. And as we highlighted, margins will be trending down closer to the longer term model of 70%, but it all depends on the mix of our hardware versus silicon.”

• Gross profit grew by 167.6% YoY to $104.5 million or 74% of revenue compared to 77.3% in the same period last year.
• Adjusted gross margin was 74.1% compared to 77.3% in the same period last year. The gross margin was below the management guide of 75% due to the higher mix of hardware revenue. Management guide for the next quarter is 74%.
• Operating margin was 0.1% compared to 17.9% in the same period last year.
• Adjusted operating margin was 34.3% compared to 24.4% in the same period last year; it was better than the guide of 32.4%, helped by operating leverage. However, operating expenses were higher than the management expectation as it continued to invest in R&D to support the strong growth and they closed a small acquisition in the quarter that also contributed to higher spending in the quarter.
• Management Q1 adjusted operating margin guide is 30.5%. The CFO said in the earnings call, “Operating expenses will grow in Q1 is largely driven by three factors, one, continued momentum in expanding our R&D resource pool across headcount and intellectual property. Two, seasonal labor, expense step ups associated with annual performance merit increases and payroll tax resets, and three, a full quarter contribution of the strategic acquisition we executed in the latter part of Q4.”
• Net income grew by 72.7% YoY to $24.7 million or 17.5% of revenue compared to 28.4% in the same period last year. Adjusted net income grew by 277.8% YoY to $66.5 million or 47.1% compared to 34.9% in the same period last year. The company benefited from a one-time tax benefit of $7.6 million in Q4.
• The vast difference between GAAP and non-GAAP net income was due to the stock-based compensation of $48.2 million or 34.2% of revenue in Q4. Once SBC naturally levels out, this company has strong enough margins to become GAAP profitable.

Strong EPS growth

Adjusted EPS grew by 208.3% YoY to $0.37 and 150% YoY to $0.30 after excluding one-time tax benefits, beating estimates by 15.8%, driven by strong operating leverage.

• Q1 adjusted EPS guide is $0.28 to $0.29, beating estimates by 14%.
• Analysts expect adjusted EPS to grow 38.5% YoY to $1.16 in 2025 and 52.6% YoY to $1.77 in 2026.

Cash Flows and Balance Sheet

The company’s cash flows are growing, helped by higher revenues.

• Q4 operating cash flow grew by 175.7% YoY to $39.7 million or 28.1% of revenue compared to 28.5% in the same period last year.
• Q4 free cash flow grew by 15.4% YoY to $15.45 million or 17.2% of revenue compared to 26.5% in the same period last year.
• Cash and marketable securities were $914.3 million compared to $886.8 million at the end of Q3 with no debt.
• Inventories increased to $43.2 million from $24.4 million at the end of Q3. Management clarified in the earnings call Q&A that they will be maintaining the current levels to support growth and the strong growth in Q3 led to the company draw down significant inventories in Q3, which led to the sequential increase in Q4.

Harlan Sur (Analyst):

Appreciate that. And on the balance sheet, inventories are up almost 80% sequentially in the December quarter. That's an all-time high for the team. I think it’s up 60% versus the average inventory level over the past four quarters. Is the significant step-up reflective of a strong multi-quarter shipment profile across the overall portfolio? Or maybe reflective of a step up in more of your board level solutions? Or is it a kind of a combination of both?

A: Jitendra Mohan (CEO and Co-Founder)

Well, if you remember, in Q3, our revenues were very strong. We were up 47% sequentially. A lot of that strength developed during the quarter. So we drew down our inventories pretty significantly in Q3. Now in Q4, we had time to build back to our more normalized level. So this this level of inventory is actually where we feel much comfortable.

We always want to be in a position to support upsize from our customers and – because most of our programs are sole sourced. But this level reflects the growth in our business now.”

Earnings Call:

Aries PCIe6 Forecasts Potential Delay on Nvidia’s GB200s

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

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

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

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

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

To stay on the topic of Astera, the following was discussed in the Q&A section:

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

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

There could be an argument made it’s unique to Astera Labs, yet we have seen something similar echoed across a few management teams in the last 4 business days, thus, it’s important to inform investors of this change in tone.

Thomas O'Malley   Barclays Bank

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

Michael Tate   CFO

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

Aries, Taurus Driving the Beat; Scorpio is Ramping Quickly

We covered Astera Lab’s product offerings in our deep dive here. The company offers a few key products that are enabling larger and faster AI clusters. For data center AI accelerators, the company was first to offer PCIe 5 switches and retimers. Growth will likely continue due to next-generation PCIe 6 products with higher average sales prices that will be released in 2025 and ramp in 2026. On PCIe6, Astera will compete against Broadcom but this is less of a concern for merchant GPUs from Nvidia, and meanwhile, Astera has indicated it’s custom silicon programs are healthy.

Increased average sales prices are being driven by CPUs, GPUs and ASICs all moving to the new PCIe 5.0 standards. Arista Labs’ Aries Retimers and PCIe 5.0 components is driving the current growth, and the company is unchallenged in this new generation of PCIe, which came to market for AI accelerators only recently with Nvidia’s H200s.

Scorpio is a new product that is expected to expand the TAM to more than $12B by 2028: Astera Labs is releasing a PCIe Gen 6 fabric switch custom designed for AI data flows with high performance per watt compared to incumbents. Rather than building a large switch, the company built a smaller device that is more efficient for high-speed signals. With Scorpio, Astera Labs is defending its dominance in PCIe5 by doubling the bandwidth at lower power requirements than the 5th generation of PCIe.

Aries products drove the company’s strong growth in 2024, along with the ramp of Taurus products in Q4. Management expects Scorpio product revenue to grow sequentially in Q1 and comprise at least 10% of total revenue for 2025.

Mike Tate, CFO, said in the earnings call, “During the quarter, we enjoyed strong revenue growth of both our Aries and Taurus Smart Cable Module products supporting both scale up and scale out PCIe and Ethernet connectivity for AI rack level configurations. For Leo CXL, and Scorpio Smart Fabric Switches, we shipped pre-production volumes as our customers work to qualify their products for production deployments later in 2025.”

Scorpio is especially promising with one analyst stating it was apparent Scorpio will drive more than 10% of revenue with ASPs being “significantly higher” than other products:

Here is what management explained: “So I think to add some color to that, again, the ASP profile of a retimer class device and a switch device tends to be very different, meaning on the switch side, we do get a significantly higher ASP. And if you look at least for the customized AI racks that are being deployed, we are essentially adding a Scorpio socket to go along with the retimer socket. And given that attach rate configuration, what we also see is that the dollar content per GPU will go up. But in general, the switch is a much bigger TAM out there. And then we get to play both in the front end with the P-Series and the back end. Back end tends to be obviously a lot more fertile in many ways because we have many GPUs talking to each other. And we benefit from having a high ASP device like the X-Series switches and them being deployed in a scale that's much more significant compared to any other products that we have released so far.”

China and Tariffs

We have covered in our editorial that semiconductor stocks face geopolitical risks from tariff threats and a possible US-China trade war in 2025, but they will also face a tougher selling climate in China as the country pushes for more domestic production towards its goal for 70% semiconductor self-sufficiency by the end of 2025. Astera Labs has 15% exposure in China, which is less of a concern for this stock.

Conclusion:

Astera is likely to see volatile price action if a Blackwell delay were to be confirmed. Its important investors have a plan and a strategy since there is an unprecedented amount of capital pointed straight at these particular AI systems (GB200s). Our plan, as it stands today, is to embrace the volatility as part of investing, hedge at times, and keep dry powder on hand with plans to buy lower. Should Blackwell ship on time (although I cannot find a component supplier who has reported yet confirming this), then we will, of course, still participate.

For the H2 vision on Astera, here is the thesis in a nutshell from our last write-up:

The products are enabling faster data speeds with PCIe5 and also increased back-end networking with PCIe6 for large AI clusters, and thus management’s commentary that this is a new growth phase holds weight. We certainly know the future for AI clusters is going to exponentially increase from 10s of thousands for AI clusters to eventually millions of AI accelerators per cluster. This is not only a GPU opportunity but also a custom silicon opportunity, as Astera Labs exclusively offered PCIe5 switches and retimers, and will now compete against Broadcom on PCIe6.

Astera Labs is technically in the lead and Broadcom is the follower in this case; but where it gets even more interesting is with the new product Scorpio. It is expected to increase the TAM by $5.0 billion with a total TAM of over $12B over the next 3 years. If we assume Astera captures 50% of the total TAM, then what we have is a stock that will remain in hypergrowth territory. If you do the math, that’s a potential 12X increase in revenue by 2028 from the $500M run rate Astera has today.

Note: Astera’s fiscal year revenue will be $676M for closer to a 10X increase, should the assumptions stated above play out.

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

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

Recommended Reading:

• Coherent FQ2: Networking Segment up 56% YoY, Keep an Eye on Co-optics for H2Coherent FQ2: Networking Segment up 56% YoY, Keep an Eye on Co-optics for H2
• AMD Q4 Earnings: Q1 Decline in Data Center, Mgmt Drops AI GuidanceAMD Q4 Earnings: Q1 Decline in Data Center, Mgmt Drops AI Guidance
• Meta Blows Past Estimates in Q4, Guides for a Soft Q1Meta Blows Past Estimates in Q4, Guides for a Soft Q1
• Amphenol: High-Performance Interconnects for the AI EcosystemAmphenol: High-Performance Interconnects for the AI Ecosystem

AMD is seeing weak price action for two reasons. First, the company is guiding for a decline QoQ from Q4 to Q1 across most segments, including the data center. Management also implied H1 of 2025 will be flat compared to H2 of 2024 in the data center. The market’s reaction is saying, “this isn’t exactly Nvidia

Second, management is dropping AI guidance from their commentary. If you’ve been following along, then you’re aware that AMD provides guidance around their expectations for AI-related GPU sales. For example, they stated this year GPUs would “exceed $5 billion” yet we started the year with commentary that GPUs would be in the range of $2 billion.

In response to AMD’s qualitative guidance, analysts would raise estimates, making it impossible for AMD to provide a surprise beat. It was a tough situation, and fairly rare to have analysts always doubling what a management had stated (on a forward basis). Therefore, I don’t disagree with what management is doing, which is creating immediate-term pain for a more equal playing field as they go ship the better architecture in the second half of 2025.

The Street thinks dropping AI guidance is hiding an issue, and communicating that there must be more weakness ahead. Almost-always, when a management team drops a key metric, it means it’s too weak to continue reporting this key metric. I believe it’s the opposite, that AMD is setting up the equal playing field before the MI350 and MI400 architecture arrive, as it’s been generally understood to be the moment when AMD can close the gap with Nvidia on some workloads.

The topic of “some workloads” is where the market will continue to get the AI story wrong, in my opinion. There is not one perfect chip, or AI system, for every single workload. It’s true that to train the trillion+ parameter models coming out of OpenAI that Nvidia’s NVL systems will be needed. However, it’s my contention that as some Big Tech workloads and enterprises seek lighter models at a reduced cost, that AMD’s new GPU architecture (coming in H2 2025) will be a strong choice.

Lastly, AMD is in pole position for edge computing, which they collectively refer to as mobile processors. Similar to data center GPUs, Nvidia will bring its sheer might and compute power as evidenced by Project Digits (the $3,000 supercomputer), yet you will you hear me continue to recognize AMD as an AI PC leader, and for lighter AI workloads at the edge (and those not looking to blow the bank), this is unlikely to change anytime soon as it’s what AMD does best.

Regarding the weak price action, overall, our firm has done quite well on this stock over the years and even recent quarters. It’s understandable if some investors are growing frustrated, yet this is not a stock we hold at a loss. We think we may be identifying a bottom, and this is key to understanding why we decide to continue to hold a stock, at times. Knox can fill you in more on Thursday on what he is seeing. If we think the stock has bottomed, then you’ll likely see us tough this one out as we are overweight Nvidia in both its stock ticker and its suppliers. We are strategically preparing for the days of “cheaper AI” which has been a thesis of ours for years; I’ve stated on Fox recently that Nvidia’s biggest threat is its gluttonous pricing power.

On the topic of custom silicon, AMD hinted its hat is in the ring. As is Nvidia. As is Broadcom, of course. As is Marvell, of course. As are a few startups. Due to the sensitive nature of IP, I foresee a scenario where each hyperscaler uses a different design company as it could expose their IP to go to one source for custom designs. We will track this as we go along.

Data Center Revenue falls short of estimates

AMD beat the revenue and EPS estimates. The company’s Q4 revenue grew by 24.1% YoY to $7.66 billion, beating estimates by 1.8%. The adjusted EPS grew by 41.6% YoY to $1.09, beating the estimates by a modest 0.4%. The revenue guidance for Q1 is $7.1 billion, representing YoY growth of 29.7% at the midpoint and beating the estimates by 1.6%.

The company’s revenue growth was led by record quarterly data center and client segment revenue. Data center revenue grew by 69% YoY and 9% QoQ to $3.86 billion. However, the revenue fell short of estimates of $4.14 billion. AI revenue exceeded $5 billion for the year 2024. Management expects strong AI and data center revenue growth in 2025. However, the earnings call lacked visibility on when AI revenue would ramp up, and guidance was missing in the earnings call.

The company’s launch of MI350 chips in mid-2025 should be a significant catalyst as the management believes that the MI350 chips have the largest generational increase in AI performance they have ever produced.

Revenue

The company’s Q4 revenue grew by 24.1% YoY to $7.66 billion, beating estimates by 1.8%. The company’s revenue growth was led by record quarterly data center and client segment revenue, partially offset by lower embedded revenue.

• Management revenue guidance for Q1 is $7.1 billion, representing YoY growth of 29.7% at the midpoint and beating the estimates by 1.6%. Management mentioned that the revenue will be down (-7%) sequentially due to seasonality.
• Analysts expect 28% growth in Q2 and 23.2% growth in Q3.

• The full-year revenue grew by 13.7% YoY to $25.8 billion.
• Looking further out, analysts expect revenue to grow 23.9% YoY in 2025 and 23.3% in 2026.
• Management expects double digit revenue and EPS growth driven by strong AI demand. The second half revenue is expected to higher than the first half due to the MI350 launch in mid-2025.

Segments

Data Center

Data Center revenue grew by 69% YoY and 9% QoQ to a record $3.86 billion. However, the data center revenue fell short of estimates of $4.14 billion and also decelerated from 122% growth in Q3. Management highlighted during the earnings call that 2024 was a significant year for the server business driven by the ramp-up of fifth-gen EYPC Turin and strong double-digit YoY growth in fourth-gen EPYC sales.

Data Center segment operating income was $1.2 billion or 30% of revenue compared to $666 million or 29% a year ago.

During the Q&A, there was a discussion that GPUs could be about $2 billion (an analyst’s words, not management – but they did not deny this number). Yet, the sticking point (as discussed) is that H1 will be flat compared to H2. Here was a response to his question as to whether the $2 billion will grow in H1: “I think you should assume that the first half of 2025 data center segment will be consistent with the second half of '24. And that's true for both businesses on the server side as well as the data center GPU side.”

Client Segment

Client segment revenue grew by 58% YoY and 23% QoQ to $2.31 billion, driven by strong demand for Ryzen desktop and mobile processors. 

Client segment operating income was $446 million or 19% of revenue compared to operating income of $55 million or 4% of revenue a year ago, driven primarily by operating leverage.

Per management, they are not concerned that inventory is building with these knockout numbers. “We don't believe there is some substantial inventory build up. We actually think that what we're seeing is very strong adoption of our new products. So on the desktop side, we saw our highest sell-out in many years, as we went through the holiday season, launching our new gaming CPUs, frankly, they have been constrained in the market, and we've continued shipping very strongly through the month of January as we are catching up with some demand there.”

Gaming Segment

Gaming segment revenue was $563 million, down (-59%) YoY, primarily due to a decrease in semi customer revenue. However, was up 22% sequentially. Gaming segment operating income was $50 million or 9% of revenue compared to $224 million or 16% a year ago.

Looking forward, management believes channel inventories have now normalized and semi-custom sales are expected to return to more historical patterns in 2025.

Embedded Segment

Embedded segment revenue was $923 million down (-13%) YoY and flat sequentially due to the slower recover in the end market. Embedded segment operating income was $362 million or 39% of revenue compared to $461 million or 44% a year ago.

Margins

The company’s margins are improving with higher data center revenue mix. Going into 2025, management expects margins to be better in the second half of the year due to higher contribution from the data center revenue and it was clarified during the earnings call Q&A. The client business expansion will be a headwind in the first half due to the higher concentration of consumer business, which has margins below the corporate average.

Toshiya Hari (Analyst)

“That's great. And then as a quick follow-up, maybe one for Jean. So you're guiding gross margin to 54% in the first quarter. I'm curious what some of the major puts and takes are and those are the things that we should be cognizant of going into Q2 and more importantly, the second half. Given your data center commentary skewed more to the second half, I would expect margins to improve in the second half. But yes, if you can kind of run through the pluses and minuses, that would be really helpful. Thank you.

Jean Hu (CFO)

Yes. Thanks for the question. You are right. Our gross margin is primarily driven by our revenue mix, I think when you look at looking to 2025, Q1 guide, not only data center continued to grow significantly year-over-year. At the same time, client business is also growing year-over-year. So overall, the revenue mix is quite consistent with the Q4. So the gross margin guide is 54%. I think for the first half, if the revenue mix is at this level, we do feel the gross margin will be consistent with 54%. But going into second half, we do believe the data center is our fastest growth driver for the company and that will drive the gross margin to step up in second half.”

• Q4 gross margin was 51% compared to 47% in the same period last year. Adjusted gross margin improved 300 bps YoY to 54% due to favourable shift of higher mix of data center and client revenues, lower gaming revenue and partially offset by lower embedded revenue. Management guide for Q1 is 54% due to the similar revenue mix as Q4.
• Operating margin was 11% compared to 6% in the same period last year. Adjusted operating margin was 26% compared to 23% in the same period last year. Operating expenses increased 23% YoY to $2.1 billion as the company invested due to the expected strong growth in AI. Management expects adjusted operating margin to be 24% in the next quarter.
• Net margin was 6% compared to 11% in the same period last year. Adjusted net margin was 23% compared to 19% in the same period last year.

EPS

The adjusted EPS grew by 41.6% YoY to $1.09, beating the estimates by a modest 0.4%. The strong growth was driven by higher data center revenue.

• Analysts expect adjusted EPS to grow 52.2% YoY in Q1 and 55.7% in Q2
• Looking further out, analysts expect adjusted EPS to grow 41.5% YoY to $4.68 in 2025 and 46.3% YoY to $6.85 in 2026.

Cash Flows and Balance Sheet

The company’s cash flows are improving driven by higher profits.

• Q4 operating cash flow grew by 241% YoY to $1.3 billion or 17% of revenue compared to 6% in the same period last year driven by higher data center revenue.
• Q4 free cash flow grew by 351% YoY to a record $1.1 billion or 14% of revenue compared to 4% in the same period last year.
• Cash and short-term investments were $5.13 billion and debt of $1.72 billion compared to $4.54 billion and $1.72 billion at the end of Q3.
• The company repurchased shares worth $256 million in Q4.

Earnings Call:

Dropping AI Revenue Guidance:

AI revenue exceeded $5 billion for the year 2024. Despite the change in tone and dropping AI guidance, there was one exchange in particular where management reiterated it will grow “strong double digits” in 2025. The details can appear confusing, as on one hand, you have H1 2025 as flat, but on the other hand, you have the growth being “strong double digits.” For anyone who follows AMD closely, it’s assumed they are referring to a strong MI350 launch in what appears to be between May-July time frame (that’s my take on the timing, considering what management has described – typically it would have been August for AMD’s cadence).

This question cut to the chase of what was on everyone’s minds – which is how management is thinking about the impact of MI350s coming given a flat H1 2025 had been discussed.

“Stacy Rasgon

Got it. Thanks. And I guess for my follow-up, maybe to follow on there, do you think your exit rate on GPUs in '25 is higher than your exit rate in '24. Are you willing to commit to that?

Lisa Su 

Absolutely. But yes, of course. It would be hard to grow strong double digits otherwise, right?”

As stated in the opening remarks, the launch of MI350 chips in mid-2025 should be a significant catalyst as the management believes the chips have the largest generational increase in AI performance they have ever produced. There was additional information provided:

“So as it relates to how data center — so the overall data center business will grow strong double digits certainly, both the server product line as well as the data center GPU product line will grow strong double digits. And from the shape of the revenue you would expect that the second half would be stronger than the first half, just given MI350 will be a catalyst for the data center GPU business. But overall, I think we are very pleased with the trajectory of the data center business in both 2024 and then going into full year 2025.”

Q1 Data Center Weakness:

Although Q4 beat on revenue, Q1 is expected to be soft with a 7% sequential decline. According to management, data center will be down just above the (-7%) total revenue sequential decline and in line with the corporate average, client and the embedded business to be down more than the total revenue decline, gaming to be down less than the total revenue decline.

The moment the stock went from being down 3% to being down 8% was when more color was provided that data center would be participating in this decline. Client is expected to be weak from Q4 to Q1, as device sales are always lower in Q1 for all consumer-device related companies (i.e., Apple, etc). Yet, seeing Gaming having stronger QoQ growth than data center is not ideal.

Here was the conversation that caused the stock to see steeper losses:

Q: Aaron Rakers (Analyst)

Yes. Thank you very much. And as a quick follow-up, just thinking about the guidance overall relative to that down 7% sequential I know you mentioned seasonality across the business segments. Are you assuming that you are down sequentially in data center in total in 1Q? And how do I frame that relative to seasonality? Thank you.

A: Lisa Su (CEO)

Yes, sure, Aaron. So let me give you some more color on the Q1 guide. So Q1 guide was down 7% sequentially, as Jean mentioned. And the way that breaks out in each of the segments assume that data center would be down just about that average, so the corporate average. We would expect the client business and the embedded business to be down more than that. Just given where seasonality is for those businesses. And then we would expect gaming business will be down a little less than that. And that's a little atypical from a seasonality standpoint, but we are coming-off of a year when there was a lot of let’s call it, inventory normalization. And now that inventory has normalized, we would expect that, that would be down a little bit less than the corporate average.”

MI350s Shipping Sooner than Expected

As with all underdogs, AMD has only a few chances to make its mark. We’ve been reporting for some time that this would be the CDNA-4 architecture with the MI350s (shipping in volume H2 2025), and if successful, then the CNDA-NEXT with the MI400s will be another opportunity (probably about a year later, H2 2026).

Here is what I’ve stated in the past:

“We will match that timeline and say we hope to see AMD be a leader in the market and in our portfolio by 2027-2028. CDNA 4 architecture is due out in 2025-2026, and is the most likely catalyst that I see today to narrow the product road map with Nvidia.”

This is what management stated about the upcoming architecture: “CDNA 4 will deliver the biggest generational leap in AI performance in our history, with a 35 times increase in AI compute performance compared to CDNA 3. The silicon has come up really well. We were running large-scale LLMs within 24 hours of receiving first silicon and validation work is progressing ahead of schedule.”

The silver lining to the earnings report is that it looks like investors will get a guide that includes some of the MI350s in the next earnings report as AMD is moving up their shipping expectations. Previously, this would have been about two quarters out. From Lisa Su: “Based on early silicon progress and the strong customer interest in the MI350 series, we now plan to sample lead customers this quarter and are on track to accelerate production shipments to mid-year.”

There were additional comments in the Q&A on this timing:

“And then the big news is on the MI350 series. So we had previously stated that we thought we would launch that in the second half of the year. And frankly, that bring-up has come up better than we expected, and there is very strong customer demand for that. So we are actually going to pull that production ramp into the middle of the year, which improves our relative competitiveness.”

By relative competitiveness, what AMD’s goal is put it into plain words, is that while Blackwell is greatly supply constrained due to the massive NVL systems, and already sold-out next year, AMD will aim to release a competitive architecture that serves the outsized demand for the B200s.

Quick Note on Custom Silicon:

There are many competitors on custom silicon, and AMD made it clear in the earnings call that they are one of them:

“And I just want to reiterate on the ASIC side, look, I think ASICs are a part of the solution, but there — I want to remind everyone, they are also a very strong part of the AMD sort of toolbox.

So we've done semi-custom solutions for a long-time. We are very involved in a number of ASIC discussions with our customers as well. And what they like to do is, they'd like to take our baseline IP and really innovate on top of that. And that's what I think differentiates our capability is that we do have all of the building blocks of CPUs, GPUs, as well as all of the networking technologies that you would need to put the solutions together.”

Conclusion:

What will get us to close our position? Price, first and foremost – which is why we sizably trimmed the position a few months ago. From what I’m hearing on the technicals side, price is doing what was expected, but you will know if this changes via our real-time trade alerts. Secondly, if the MI350s are too weak to take the overflow in capex that Blackwell supply cannot fulfill. If consumer and tariffs create more impact than currently priced in. We always have a plan.

With that said, I will be watching with keen interest the launch of the MI350s. If this release benchmarks “close enough” to the B200s then AMD investors will be rewarded. If the benchmarks are too far apart, the proverbial David (Lisa Su and team) will have missed a critical window for striking with her slingshot. Despite what benchmarks say, we only have to watch how Big Tech capex funnels to truly know how AI accelerators are being received; this is a proxy and substantial (unprecedented) flow of capital that overwhelmingly matters more than what analysts, industry experts, or even management teams say or predict. Big Tech companies are at the forefront and they will make the ultimate decision.

Right now, that capex is pointed directly at Nvidia, which was the basis of our Q1 webinar. We will see come late Spring/early Summer if AMD can pry some of the roughly $300 billion loose. The market is giving up, but that sometimes becomes precisely the moment when a stock bottoms.

Before I conclude, keep an eye on the Client segment as it’s abundantly clear who is in the lead with AI PCs. Nvidia will do quite well here too, but AMD’s numbers are communicating something very important about their lead in edge AI.

Royston Roche, Equity Analyst for the I/O Fund, 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 AMD at the time of writing and may own stocks pictured in the charts.

Recommended Reading:

• AMD Q4 Earnings Preview: The Bottom May Be Near
• Microsoft FQ2 Earnings: Soft revenue guidance
• TSMC Q4 2024 Earnings: Record Profit Led by Robust AI Demand
• Micron Q1: Data Center Revenue Surges 40% QoQ but Consumer Weak