Analyst(s): Brendan Burke
Publication Date: November 6, 2026
Astera Labs is expanding its connectivity portfolio to include the fabric, memory, and software layers of AI infrastructure. Scorpio adoption and Orion’s reference architecture broaden its role in improving inference economics, with customer qualification and deployment timing shaping execution risk.
What Is Covered in This Article:
- Astera Labs’ Q3 FY 2026 results
- Scorpio expands scale-up fabric adoption
- Memory and optics broaden connectivity content
- Orion adds software, licensing, and 160-GPU rack scale-up opportunities
- Guidance and Final Thoughts
The News: Astera Labs (Nasdaq: ALAB) reported Q3 FY 2026 results for the quarter ended September 30, 2026. Revenue reached $584.9 million, up 154% year-over-year (YoY), compared with Wall Street consensus of $548.4 million. Revenue exceeded the upper end of its previously issued $540 million to $560 million range by $24.9 million, or 4.45%. Non-GAAP operating income increased 172.14% YoY to $261.65 million, and operating margin rose 300 basis points YoY to 44.7%. Non-GAAP net income increased 175.9% YoY to $243.4 million, and diluted earnings per share (EPS) rose to $1.32 from $0.49 in the prior year.
“Design activity accelerated with wins across a broad set of programs and applications at multiple customers,” said Jitendra Mohan, chief executive officer of Astera Labs.
Astera Labs Q3 FY 2026: Orion and Scorpio Expand Infrastructure Content Opportunity
Analyst Take: Astera Labs’ strategic opportunity is expanding as AI infrastructure buyers focus on accelerator utilization and inference cost. Fabric switches, memory controllers, and connectivity software give the company more ways to address those constraints within the same system. More than 10 active custom silicon engagements add another route into customer-specific architectures. The strategic test is whether a broader portfolio converts into repeat deployments across customers rather than deeper dependence on a few large platform ramps.
Scorpio Expands Scale-Up Fabric Adoption
Scorpio’s adoption across additional platforms strengthens Astera Labs’ position in scale-up networks, where accelerators communicate within a tightly connected compute domain. X-Series shipped to two customers during the quarter, while P-Series deployments included two US hyperscalers. The 320-lane X-Series switch supports larger, flatter network topologies that reduce the number of switching hops and associated communication delays. Its Hypercast and in-network compute engines move selected collective operations into the fabric, with Astera Labs reporting up to twice the collective performance.
Existing configurations can already contain more than $1,000 of Astera Labs fabric switch and copper signal conditioning content per accelerator, although that figure is not a portfolio-wide average. Buyers should assess fabric performance alongside accelerator specifications because communication efficiency can determine how much purchased compute becomes usable inference capacity.
Memory and Optics Broaden Connectivity Content
Leo and Aries 7 extend Astera Labs’ opportunity into memory access and the physical links connecting larger AI systems. Leo X-Series pairs with Scorpio to create fabric-attached memory for KV caches, addressing a capacity constraint in long-context inference. Company internal testing showed up to 62% faster time-to-first-token and up to 22% more tokens per second, results that require validation against each customer’s workload. Leo 2 doubles memory bandwidth and capacity versus its predecessor, while qualifications across the Leo portfolio include two hyperscalers. Aries 7 extends Peripheral Component Interconnect Express (PCIe) signal conditioning across copper and optical links, with optical reach exceeding 50 meters and a shared footprint allowing retimer or redriver selection without board redesign. The competitive opportunity lies in capturing more memory and connectivity content per system, with workload-specific performance and qualification providing the evidence needed for adoption.
Orion Adds Software and Licensing Opportunities
Orion turns Astera Labs’ component portfolio into a validated, compute-agnostic reference architecture that allows customers to decouple accelerator and CPU selection in AI buildouts. Delivered across six production-ready reference designs—spanning compute, fabric, and memory trays, along with fabric and link boards—Orion supports configurations scaling up to 80 GPUs per rack in a single PCIe domain and up to 160 accelerators per pod across racks. This high-radix 320-lane PCIe switch fabric and composable architecture reduce integration complexity while improving token-per-second economics and addressing KV cache limits via fabric-attached CXL memory. The commercial model combines baseboards, COSMOS software, and reference-design royalties through OEM/ODM partners like Hyve Solutions. COSMOS extends beyond device diagnostics into rack and fabric management, creating a software monetization point alongside hardware. Orion’s strategic value depends on whether validated integration and management software can earn recurring platform adoption in AI cloud and enterprise markets with varied workloads and deployment requirements.
Guidance and Final Thoughts
Astera Labs guided Q4 FY 2026 revenue to $730 million to $750 million, with the $740 million midpoint approximately 12.3% above Wall Street consensus of $658.7 million and 26.5% above Q3 FY 2026 revenue. Non-GAAP gross margin is expected to reach approximately 72%, with operating expenses of $194 million to $200 million, operating margin of approximately 45%, and diluted EPS of $1.62 to $1.66. The outlook reflects continued Scorpio expansion alongside increasing Aries and Taurus adoption, while operating expenses are projected to rise approximately 22% sequentially at the midpoint as Astera Labs invests in a broader connectivity portfolio. Sustaining this revenue trajectory will increasingly depend on new customer platforms as initial Scorpio X-Series deployments mature.
Astera Labs’ next growth phase depends on expanding its share of AI infrastructure spending beyond fabric switching into memory connectivity, optical links, and integrated rack-level systems. Orion’s Q1 FY 2027 availability through partners like Hyve provides an early opportunity to demonstrate the value of validated architectures and COSMOS software, while Leo volume shipments in mid-FY 2027 and initial optical revenue later in the year could broaden content per deployment. However, staggered qualification schedules and custom-silicon production targeted for FY 2028 mean these opportunities will contribute at different stages, leaving near-term growth concentrated in established products. If Astera Labs can convert customer qualifications into production ramps while maintaining its approximately 72% gross margin profile, then its expanding portfolio could support a more diversified revenue base as AI infrastructure architectures evolve.
See the full press release on Astera Labs’ Q3 FY 2026 financial results on the company website.
Declaration of generative AI and AI-assisted technologies in the writing process: This content has been generated with the support of artificial intelligence technologies. Due to the fast pace of content creation and the continuous evolution of data and information, The Futurum Group and its analysts strive to ensure the accuracy and factual integrity of the information presented. However, the opinions and interpretations expressed in this content reflect those of the individual author/analyst. The Futurum Group makes no guarantees regarding the completeness, accuracy, or reliability of any information contained herein. Readers are encouraged to verify facts independently and consult relevant sources for further clarification.
Disclosure: Futurum is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.
Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of Futurum as a whole.
Read the full Futurum Group Disclosure.
Other Insights From Futurum:
Astera Labs Q2 FY 2026 Earnings Pull Scorpio Switch Leadership Forward
Astera Labs Orion Reimagines the XPU Rack as a Compute-Agnostic Platform
Astera Labs Aries 7 Goes Optical. Can PCIe 7 Outgrow the Retimer Plateau?
Author Information
Brendan is Research Director, Semiconductors, Supply Chain, and Emerging Tech. He advises clients on strategic initiatives and leads the Futurum Semiconductors Practice. He is an experienced tech industry analyst who has guided tech leaders in identifying market opportunities spanning edge processors, generative AI applications, and hyperscale data centers.
Before joining Futurum, Brendan consulted with global AI leaders and served as a Senior Analyst in Emerging Technology Research at PitchBook. At PitchBook, he developed market intelligence tools for AI, highlighted by one of the industry’s most comprehensive AI semiconductor market landscapes encompassing both public and private companies. He has advised Fortune 100 tech giants, growth-stage innovators, global investors, and leading market research firms. Before PitchBook, he led research teams in tech investment banking and market research.
Brendan is based in Seattle, Washington. He has a Bachelor of Arts Degree from Amherst College.
