Astera Labs is navigating an important rite of passage that many post-IPO hypergrowth companies stumble through, which is to offer a consistent growth trajectory. Once the excitement of an IPO fades, most tech companies cannot sustain the growth the private sector primed the company for ahead of listing on the public markets.
Astera is bucking this trend, as the company was expected to report 8% QoQ growth and instead reported 14% QoQ growth, which resulted in an official 1.6 percentage points acceleration on a YoY basis from 91.8% YoY growth last quarter to 93.4% growth this quarter. Looking ahead, Astera is offering a guide that indicates QoQ growth of 16.7% and YoY growth of 87.6% for revenue of $360 million at the midpoint. This handily beat forward expectations for revenue of $310.1 million next quarter.
As we look toward the second half of the year, management offered strong commentary that suggest growth will continue: “As we look to the second half of 2026, robust demand reflects secular AI infrastructure spending, deep customer partnerships and expansion towards higher-value solutions within our portfolio. […] As a result, we expect strong revenue growth to continue through 2026 and into 2027, driven by the proliferation of AI fabrics and the industry's transition to PCIe 6, 800 gig and 1.6T Ethernet connectivity.”
Perhaps most notable is that Astera delivered a strong quarter even after the stock declined roughly 55% peak-to-trough from September through March, reflecting a disconnect between sentiment and fundamentals.
Scorpio-X 320 Lane Smart Fabric Switch
Positioning:
In this evening’s print, Astera also announced the Scorpio X-Series 320 Lane Smart Fabric Switch, which is the largest open, memory semantic fabric switch on the market with 5.12 TB/s bidirectional bandwidth in a single ASIC. The 320 Lane variant offers 16 lanes per device and 20 accelerators per switch, which is roughly “2x the radix in a single hop,” which means twice the number of GPUs are connected on the same switch. With Scorpio-X, only one switch is needed for 320 GPUs, and fewer switch hops means lower latency.
Astera differentiates itself from Broadcom’s Ethernet switch Tomahawk 6 and Nvidia’s NVSwitch by providing an open PCIe-based fabric for CPUs, NICs and storage with the P-Series and improving accelerator-to-accelerator performance specifically around memory sharing with the X-series.
The X-Series is timed to the scale-up networking opportunity and the inference market. Mixture of Experts (MoE) inference is a steady stream of tasks, which requires very fast accelerator-to-accelerator communication. As discussed in the call, MoE requires frequent routing of tokens and data across expert models, which places more emphasis on the scale-up fabric. Astera Labs is uniquely positioned to enable GPUs and AI accelerators to communicate more efficiently across PCIe, especially when it comes to direct memory access.
Here is what was stated in the opening remarks:
“Scorpio X-Series portfolio now supports up to 320 lanes for high radix scale-up networking, and Scorpio P-Series PCIe 6 portfolio now spans 32 to 320 lanes for diverse system topologies, making it the broadest in the industry.
Our new flagship Scorpio X-Series 320 lane has been purpose-built to maximize AI economics by leveraging hardware-accelerated hypercast and in-network compute engines to boost collective operations by up to 2x. In-network compute offloads critical accelerator to accelerator communication and computation directly onto the switch, dramatically reducing the networking overhead during large-scale training and inference.”
This is a significant shift as it brings the math operations inside the switch instead of the GPUs, which Astera is referring to as “in-network compute.” Hypercast refers to handling operations inside the switch, which reduces the networking overhead associated with GPU-to-GPU coordination. The result for inference tasks is more tokens per dollar as Scorpio-X removes the need for GPUs to wait on other GPUs during MoE and agentic workloads.
It's important to double-click on the memory-semantic piece. Astera's fabric lets accelerators access each other's memory directly like a single unified memory pool, eliminating the overhead of translating data into network packets. This is important for AI workloads, and especially MoE inference, which depend on constant sharing of weights, activations, KV cache, etc., across accelerators. Per the press release: “Its memory-semantic connectivity enables accelerators to access fabric resources through native load/store operations, eliminating software overhead and improving fabric efficiency at scale.”
Astera’s X-Series offers communication across mixed architectures (both GPUs and ASICs) but also solves for memory sharing – both are key as we move into the inference market.
Economics:
We’ve covered the X-Series for about a year in our post-earnings analyses. For investors, some of the most important takeaways is that the Scorpio product is expected to increase from 15% of product mix at the end of CY25 to 50% of product mix by the end of CY26. Although the P-Series is driving the current growth, the X-Series will be the higher mix as we exit the year – which means this ramp is second-half weighted.
“Given the size of the opportunity and the associated dollar content, we would expect to see that Scorpio will become our largest product line by the end of the year, which is strong performance for a product line that was only 15% of total company revenue last year. And as we go throughout the year, I would expect to see X-Series revenue exceeding P-Series.”
Another point for investors is the average sales prices will increase from the X-Series. Here is what was stated on the call:
“Yes. So in general, what I would say is the bigger the switch, the higher the ASP. That's the way industry works. But also, please keep in mind is that these switches are more like AI fabric class device, which are a lot more than just the number of lanes, right? […]So when it comes to ASP, obviously, it's a combination of how — what features are enabled and not just based on the port count. But we do see that our content continue to increase. And to that standpoint, we are expecting and going forward with the design wins we have, over $1,000 worth of content per accelerator.”
Future Product Roadmap for 2027-2028
Optical Opportunity:
Astera’s optical roadmap is an extension of the company’s ability to offer end-to-end PCIe over optics for GPU clusters. As racks grow into larger pods, cable length and signal integrity become constraints. Astera has stated at a recent investor’s event that optical becomes necessary at higher data rates (which is also general consensus).
Last October, Astera acquired a scale-up photonics company to offer optical scale-up interconnects. On the earnings call, it was shared that near-packaged optics will roll-out first following this acquisition in 2027, which is a bridge solution while co-packaged optics may take longer than the market cares to wait.
Here is what was stated on the call:
“For us, in terms of time line, what we believe is that the NPO-based opportunities, or the near package optics, would be the first one to ramp, and that will start happening in 2027. We will also be ramping our pluggable connector technologies for CPO, mostly for scale-out next year, 2027, with more of the mainstream deployments for CPO happening in the 2028 time frame.”
NVLink Fusion Opportunity:
Notably, Astera Labs offers connectivity solutions for hybrid AI racks. This widens Astera’s content opportunity beyond UALink as it provides an additional path to scale-up AI fabrics by offering a bridging solution for GPUs and custom silicon. In some cases, when NVLink is chosen, Astera will still be a key supplier for connectivity solutions.
“Clearly, an area that we see tremendous opportunity for us going forward is the custom solutions under which we are developing the NVLink Fusion type of devices. And this actually is proving to be pretty interesting. We do have several opportunities. We're very deep in engagement for an initial design win in collaboration with NVIDIA and then a hyperscaler. So that project is going well. So we do expect that to start contributing revenue in 2027 as some of the GPUs that are designed for this kind of use case, which is called as a hybrid rack situation, where the GPU or the XPU still talks native protocols, which could be a protocol like PCIe or UALink and others. But then when they need to leverage and cross over and talk to an NVLink type of ecosystem, then they would need a product that's based on NVLink Fusion that we are developing.”
CXL Opportunity:
CXL is a longer-term opportunity for Astera Labs, and will extend Astera’s content opportunity (again) to include memory pooling and connectivity. This provides more direct exposure to the memory side of the AI buildout rather than only the accelerator interconnect. Here was the update for the call, including a newer customer win that could help with KV cache offload: “Finally, our LEO memory controller is on track for an early ramp of CXL attached memory with Microsoft Azure M-Series virtual machines. And during the quarter, we captured a new custom design win for a KV Cache offload application with shipments expected in 2027.”
Note on UALink:
We’ve written in the past that UALink as a scale-up fabric is expected to go head-to-head with Ethernet Scale-Up Networking (ESUN). In the past, when there are ESUN announcements, ALAB’s stock reacts negatively. However, that assumes a zero-sum outcome, whereas it’s more likely scale-up sees a mix of both UALink and ESUN.
The quick refresher is that ESUN is attempting to make Ethernet work for scale-up whereas UALink was built from scratch for scale-up. The primary benefit ESUN offers is to move quicker than UALink (as discussed above, ALAB is saying it’ll be 2027 for UALink to be fully deployed). However, in the meantime, Astera’s PCIe solutions are in high demand and deployable now.
Even if ESUN moves faster commercially, there is a performance gap that helps to ensure that Astera’s positioning with PCIe/CXL remains intact. That performance gap is best described as the low latency required for what are the most in-demand AI workloads today – those that require memory pooling and GPU-to-GPU communication.
For more information, read our previous analysis here.previous analysis here.
Financials
By Royston Roche
Revenue Accelerates to 93.4% YoY
Astera Labs reported Q1 2026 revenue of $308.4 million, beating estimates by 5.5%. Growth continued at a robust pace on a YoY basis, with revenue up 93.4% YoY and accelerating 1.6 percentage points from 91.8% growth in the previous quarter. On a sequential basis, revenue grew 14.0% QoQ from $270.6 million in Q4 2025.
Aries product revenue grew strongly in Q1 2026, with PCIe Gen 6 solutions for both scale-out and scale-up signal conditioning driving solid adoption. Management noted that PCIe Gen 6 revenue across AI fabric and signal conditioning contributed more than one-third of total revenue in the quarter — a significant milestone reflecting the accelerating industry transition to Gen 6.
The Scorpio product family also performed well in Q1, driven by strong demand for PCIe Gen 6 switching applications and continued expansion of designs across various platforms. During the quarter, Scorpio X-Series products began shipping in initial production volumes. Management expects Scorpio X-Series shipments to increase in Q2, along with initial shipments of the new Scorpio X 320 lane product and then ramp to full volume production in the second half of 2026.
Taurus product family continued to deliver solid results in Q1 2026, driven by broad adoption of Active Electrical Cable (AEC) to extend reach in both AI and general-purpose compute platforms.
Leo's CXL memory expansion products continue to advance, with management highlighting an early production ramp of CXL-attached memory with Microsoft Azure M-Series virtual machines and a new custom design win for a KV Cache offload application with shipments expected in 2027.
Management guided strong Q2 revenue guidance of $355 million to $365 million, implying a YoY growth of 87.6% and 16.7% QoQ at the midpoint, beating estimates by 16.1%. Aries revenue growth is expected to be driven by continued strong adoption of PCIe 6 across AI platforms, supporting both scale-up and scale-out connectivity. Taurus growth is expected to be driven by increased volumes for AI scale-out connectivity. And in AI fabric, management expects robust growth driven by the continued early-stage ramp of the Scorpio X-Series products for large-scale XPU clustering applications as well as continued growth in the P-Series solutions and customized GPU platforms.
Margins Beat Guidance
Astera Labs delivered impressive gross margin performance in Q1 2026, with GAAP gross margin coming in at 76.3%, comfortably ahead of the 74% guidance. This compares to 75.6% in Q4 2025 and 74.9% in the same period last year, a sequential expansion of 70 basis points and 140 basis points YoY, primarily due to favorable product mix. Adjusted gross margin improved 150 basis points YoY to 76.4%.
Management has guided adjusted gross margin to be lower at 73% for the next quarter, primarily due to the estimated 200 basis point noncash impact related to a recently executed warrant agreement with one of its customers.
GAAP operating margin improved 13 percentage points YoY to 20.1%. While adjusted operating margin improved 2.5 percentage points YoY to 36.2% primarily due to operating leverage and beat the guidance of 34.5%.
Q1 2026 adjusted net income grew by 84.7% YoY to $110.7 million or 35.7% of revenue compared to 37.4% of revenue in the same period last year.
Adjusted EPS grew by 84.8%
Q1 adjusted EPS grew by 84.8% YoY to $0.61, beating estimates by 13.5% primarily due to operating leverage. GAAP EPS growth was even stronger as it grew by 144.4% YoY to $0.44 and beating estimates by 26.5%.
Management also provided a strong EPS guide for the next quarter. GAAP EPS guide is $0.45 at the midpoint, up 55.2% YoY and beat estimates by 37.6%. Adjusted EPS guide is $0.69 at the midpoint, up 56.8% YoY and beat estimates by 25.5%.
Cash Flow and Balance Sheet
The company’s cash flows were strong primarily due to higher profits.
- Q1 operating cash flow was $74.6 million or 24.2% of revenue compared to a mere $10.5 million or 6.6% of revenue in the same period last year.
- Q1 free cash flow was $67 million or 21.7% of revenue compared to $5.97 million or 3.7% of revenue in the same period last year.
- The company maintains a robust balance sheet with cash & marketable securities of $1.18 billion and no debt.
- Inventories rose 2% QoQ to $60.2 million.
Conclusion:
Astera Labs is expanding their product road map well beyond selling PCIe retimers, and is now solving serious bottlenecks for the incoming AI inference market. Inference workloads are more memory-intensive and will see ongoing, exponential accelerator-to-accelerator communication, not to mention the critical importance of shared memory access.
Astera’s role is becoming more strategic as the company has multiple paths to increase content through Scorpio-X scale-up switching, both UALink and NVLink Fusion content opportunities, CXL memory pooling, and they’re prepared for the optical transition – whew, that’s a lot. Near-term volatility could persist as the market debates protocol winners, but one thing is for certain – AI workloads are becoming more complex. Astera is on the front lines of solving that complexity.
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 ALAB at the time of writing and may own stocks pictured in the charts.
Recommended Reading: