Penguin Solutions, Inc.

Penguin Solutions is the AI Factory Platform Company. We design, build, deploy, and manage next-generation data centers for enterprises, sovereign AI initiatives, and neocloud providers.

With deep design expertise at the intersection of data center AI infrastructure and memory solutions, our Full-Stack AI Factory Platform combines differentiated infrastructure software, advanced memory, compute systems, end-to-end services, and industry-leading partner technologies to help customers accelerate deployment, optimize token economics, and maximize the return on their AI investments.

As of the end of fiscal 2026, Penguin Solutions employed approximately 2,800 employees worldwide, with most located in the United States, China, and Malaysia. We believe that our employees are the cornerstone of our success. To support their efforts, we aim to provide inclusive and equitable workplaces maintained through ongoing intentional actions.

On October 15, 2024, we changed our corporate name from “SMART Global Holdings, Inc.” to “Penguin Solutions, Inc.” and changed our Nasdaq Global Select Market ticker symbol from “SGH” to “PENG.” The changes to the corporate name and ticker symbol did not have any impact on our legal entity structure, financial statements or previously reported financial information.

On June 30, 2025, we consummated the U.S. Domestication of the parent company of our corporate group from the Cayman Islands to the State of Delaware in the United States. For more information about the U.S. Domestication, see “About This Annual Report” above.

Business Segments

Penguin Solutions has three business segments: Advanced Computing, Integrated Memory, and Optimized LED. Each brings specialized expertise, innovative technologies, and a commitment to delivering high-quality products and services that meet evolving customer needs.

Our Advanced Computing and Integrated Memory businesses are central to our AI Factory Platform strategy, combining expertise in AI infrastructure, advanced memory technologies, and end-to-end services to help customers deploy, scale, and optimize next-generation AI data centers.

Our Optimized LED segment, operating as Cree LED, serves a distinct market with a dedicated strategy focused on advancing high-performance LED technologies and delivering innovative solutions to customers worldwide.

On November 29, 2023, we completed the divestiture of an 81% interest in SMART Modular Technologies do Brasil – Indústria e Comércio de Componentes Ltda. (“SMART Brazil”) to Lexar Europe B.V. (“Lexar Europe”), an affiliate of Shenzhen Longsys Electronics Co., Ltd. During 2026, we sold our remaining 19% interest in SMART Brazil, subsequently renamed Zilia Technologies, completing our divestiture of the business.

Presentation of SMART Brazil as Discontinued Operations: In accordance with authoritative guidance under U.S. GAAP, we have presented the balance sheets, results of operations and cash flows of SMART Brazil operations in this Annual Report, including in the accompanying consolidated financial statements and notes, as discontinued operations for all periods presented. The SMART Brazil operations were previously reported as part of our Integrated Memory segment. Unless otherwise noted, discussion within this Annual Report relates solely to our continuing operations and excludes the SMART Brazil operations. See “PART II – Item 8. Financial Statements and Supplementary Data – Notes to Consolidated Financial Statements – Divestiture of SMART Brazil” and “PART II – Item 8. Financial Statements and Supplementary Data – Notes to Consolidated Financial Statements – Cash and Investments – Non-marketable Equity Investments – Zilia Technologies.”

5

Our Products and Services

Advanced Computing

Our Advanced Computing segment delivers AI infrastructure solutions that help customers design, build, deploy, and manage next-generation AI data centers. Through our Full-Stack AI Factory Platform, we combine advanced compute systems, differentiated infrastructure software, and end-to-end services with industry-leading partner technologies to help enterprises, sovereign AI initiatives, and neocloud providers accelerate deployment, optimize performance, and improve token economics.

Our portfolio includes AI and high-performance computing (“HPC”) solutions offered under the Penguin Solutions® and Penguin Computing® brands, including our ClusterWareAI™ AI factory operating system software, MemoryAI™ data center and AI inference solutions, ComputeAI™ GPU and CPU computing systems, and OriginAI® reference architectures. Our services span the AI infrastructure lifecycle, from design and deployment to ongoing optimization and managed operations.

We also offer high-availability and fault-tolerant computing solutions under our Stratus® brand, supporting customers that require continuous availability and reliable operations in data center and edge environments.

Our Advanced Computing solutions serve enterprises, hyperscalers, sovereign AI initiatives, and neocloud providers. We help customers deploy and scale next-generation AI infrastructure to support increasingly demanding AI workloads.

Our Advanced Computing segment had net sales of $558.8 million, $648.4 million, and $554.6 million in 2026, 2025, and 2024, respectively.

AI and HPC Data Centers

With more than 25 years of experience in AI and HPC infrastructure, Penguin Solutions designs, builds, deploys, and manages advanced computing environments that support demanding, data-intensive workloads. Our deep technical expertise and relationships with industry-leading compute, networking, storage, and software partners enable us to deliver scalable, high-performance AI infrastructure tailored to our customers’ needs.

Our Full-Stack AI Factory Platform combines OEM-like product innovation with system integrator-like end-to-end services and industry-leading partner technologies. Bringing together differentiated infrastructure software, advanced memory solutions, compute systems, and services, our platform helps enterprises, sovereign AI initiatives, and neocloud providers accelerate deployment, optimize performance, and scale AI infrastructure. As AI shifts toward inference-centric workloads and agentic AI, our integrated approach is designed to improve token economics and maximize the return on AI investments.

Our platform is built on five core elements:

OriginAI

OriginAI provides pre-validated AI infrastructure architectures designed to accelerate deployment and deliver predictable performance for GPU clusters ranging from hundreds to thousands of GPUs.

These architectures combine advanced compute, networking, storage, and software technologies with validated configurations to streamline infrastructure design, integration, and deployment. By reducing the time and effort required to configure and validate complex AI environments, OriginAI helps customers accelerate deployment, scale infrastructure efficiently, and achieve more predictable operational performance.

MemoryAI

Our MemoryAI portfolio is designed to address memory capacity, bandwidth, and efficiency challenges that increasingly constrain AI infrastructure performance, particularly for inference workloads. Our initial solutions focus on Compute Express Link or CXL-based memory technologies, including Key Value Cache memory expansion, to improve memory utilization and support more efficient inference operations. We are developing

6

additional capabilities intended to help customers improve GPU efficiency, optimize token economics, and maximize the return on their AI infrastructure investments.

ComputeAI

Our ComputeAI portfolio includes advanced compute systems designed to support demanding AI and HPC workloads. These systems provide the high-performance computing foundation for AI infrastructure, combining accelerated computing technologies with flexible system configurations to meet diverse workload and deployment requirements.

Our compute solutions support a broad range of applications, from traditional AI and HPC model training to increasingly demanding inference workloads. By integrating advanced processors, accelerators, memory, and networking technologies, ComputeAI enables customers to deploy and scale high-performance computing environments tailored to their operational needs.

ClusterWareAI

ClusterWareAI, our hardware-agnostic AI factory operating system software, provides centralized management, monitoring, automation and is designed to help optimize AI infrastructure. Its capabilities include cluster deployment, observability, GPU health monitoring, policy-based management, and automated remediation designed to help customers improve infrastructure reliability, operational efficiency, and GPU utilization.

We are extending these capabilities through our Agentic Fabric development, designed to enable intelligent orchestration and coordinated operational workflows across compute, networking, storage, and workloads.

End-to-End AI Infrastructure Services

Our end-to-end services span the AI infrastructure lifecycle, from initial design and engineering through integration, deployment, optimization, and ongoing managed operations. We work closely with customers to develop infrastructure solutions tailored to their technical requirements, business objectives, and operational environments.

Our services include AI infrastructure design, system integration, deployment, cluster validation, performance optimization, and ongoing infrastructure management. Through our managed services capabilities, we provide continuous monitoring, operational support, and infrastructure optimization to help customers maintain system availability, improve operational efficiency, and scale their AI environments.

Together, our integrated products and services enable customers to accelerate AI infrastructure deployment, address operational challenges, and maximize the value of their AI investments throughout the AI factory lifecycle.

Fault-Tolerant Computing

Our Stratus brand products are designed to ensure the continuous availability of customers’ critical data and applications in data centers and edge locations. Our Stratus ztC® platforms deliver fault tolerance that is simple to install, interoperable with existing systems, manageable by time-constrained IT or operational technology (“OT”) teams, and protected from interruptions and threats. For over 40 years, our fault-tolerant computing solutions have provided reliable and redundant computing that supports Fortune 500 companies as well as small- to medium-sized businesses in securely and remotely turning critical data into actionable intelligence. As a result, IT and OT professionals across financial services, oil and gas, transportation, healthcare, retail and discrete manufacturing rely on our platforms and services to mitigate operational, financial, and reputational risk and support “always on” availability.

Our solutions include the Stratus ztC Endurance® compute platforms which are designed to deliver 99.99999% availability. The Stratus ztC Endurance and its Automated Uptime Layer with Smart Exchange™ capability delivers intelligent, predictive fault-tolerant computing within an easily serviced modular architecture. These platforms evolve from the built-in fault tolerance and proactive health monitoring found in our fully integrated and continuously available Stratus ftServer® platforms. Additionally, the Stratus ftServer V series platforms deliver advanced availability, performance and scalability for high-volume, high-risk transaction processing. The Stratus ztC Edge® platform brings high availability and fault tolerance to IT constrained edge environments, such as retail locations and manufacturing sites, in a secure, highly automated Class 1, Div 2-certified computing platform.

7

Lastly, our Stratus everRun® software protects virtualized workloads on third-party hardware. Our Stratus platforms and software are backed by expert services professionals dedicated to customer success.

Discontinued Computing Solutions

In the fourth quarter of fiscal 2026, we fully discontinued our Penguin Edge product portfolio, which previously combined our SMART Embedded Computing™ (“SMART EC”) and SMART Wireless Computing™ (“SMART Wireless”) brands under the unified Penguin Edge brand. The decision reflects our strategy to focus resources on higher-growth opportunities. See “PART II – Item 8. Financial Statements and Supplementary Data – Notes to Consolidated Financial Statements – Intangible Assets and Goodwill.”

Integrated Memory

Our Integrated Memory segment designs, develops, and delivers advanced memory and storage solutions that address performance, capacity, and reliability requirements across AI infrastructure, high-performance computing, networking, and enterprise applications.

As AI adoption accelerates and workloads become increasingly memory-intensive, our Integrated Memory solutions help customers address growing memory capacity and performance demands. Our portfolio includes specialty memory and storage products sold under our SMART Modular Technologies® product brand, including dynamic random access memory (“DRAM”) modules, solid-state and flash storage solutions, and CXL-based memory expansion technologies designed to improve memory capacity, bandwidth, and system efficiency.

Our Integrated Memory segment also provides SMART Supply Chain Services, offering customized supply chain planning and execution solutions that help customers improve operational efficiency, reduce costs, and increase productivity. Our Integrated Memory segment had net sales of $924.0 million, $464.2 million and $356.4 million in 2026, 2025 and 2024, respectively.

Memory Products

Our memory solutions have supported customers across diverse industries and demanding applications, including data centers, networking, industrial automation, rugged environments, and emerging AI and cloud infrastructure, for more than 30 years. Through innovative design, development, and advanced packaging of specialty integrated memory and storage solutions, we address complex memory challenges across a broad range of applications. Our portfolio spans leading-edge memory technologies, standard DRAM and flash storage products, and ruggedized and customized memory and storage solutions designed to meet specialized performance, capacity, and reliability requirements.

We collaborate closely with global original equipment manufacturers (“OEMs”) throughout their design processes and across multiple projects to develop solutions for demanding applications with differentiated requirements. These include unique form factors, customized firmware, higher memory density, lower power consumption, and enhanced durability and reliability compared to standard solutions. We are a primary supplier of longer-lifecycle solutions to OEM customers across industrial, government, networking and communications, enterprise storage and computing, and other vertical markets. We offer an extensive portfolio of memory products in standard and rugged form factors.

DRAM and Specialty Memory Solutions

Our offerings include an extensive lineup of DRAM modules across a variety of technologies, including legacy synchronous DRAM, double data rate (“DDR”), DDR2, DDR3, and leading-edge, high-performance DDR4 and DDR5 DRAM devices. These technologies are incorporated into standard memory, enterprise memory, and hybrid integrated memory solutions in both standard and rugged formats.

Our DRAM modules encompass a broad range of form factors and functions, including dual in-line memory modules (“DIMMs”), differential DIMMs, load-reduced DIMMs, registered DIMMs, unbuffered DIMMs, small-outline DIMMs, and mini-DIMMs for industrial, government, networking and communications, enterprise storage and computing, and other vertical markets.

These memory modules are available in configurations of up to 288 pins and densities of up to 256 gigabytes. We utilize advanced printed circuit board and device packaging and stacking technologies to achieve cost-effective,

8

high-density solutions. Our products are designed to meet the quality requirements of enterprise-class systems and the stringent specifications of various high-speed applications.

Zefr® High-Reliability Memory

Our Integrated Memory segment also offers Zero Failure Rate (“Zefr”) memory modules designed to support mission-critical applications and HPC platforms that process massive amounts of data. Zefr memory modules undergo a rigorous proprietary screening process, performed on OEM memory modules or SMART Modular Technologies memory modules, to deliver ultra-high reliability for demanding workloads. Zefr is designed to maximize memory uptime and system availability, including enhancing the reliability and uptime of our Stratus ztC Endurance platform.

CXL Memory Expansion

Additionally, we support leading-edge and emerging interconnect standards such as CXL, an industry-standard, open protocol that enables high-speed, low-latency connectivity and memory expansion between processors and compatible devices. Our CXL-based memory expansion solutions are designed to address growing memory capacity requirements for AI, HPC, and other memory-intensive workloads. SMART CXL Memory is available as a memory expansion add-in card (“AIC”) in our Penguin Solutions Altus® AMD EPYC™-based servers, providing expanded memory capacity for demanding applications.

Flash Memory and Storage

We also design and manufacture embedded and removable flash memory products in a variety of form factors and capacities, incorporated into storage and hybrid integrated memory solutions in standard and rugged formats. Our flash memory products include solid-state drives (“SSDs”), Serial Advanced Technology Attachment (“SATA”) and PCIe NVMe products in 2.5-inch enclosures, M.2, EDSFF, and other module form factors. We also offer flash component products such as embedded MultiMediaCard, embedded and removable USB products, and SD and microSD card configurations. Our flash memory capabilities include customized firmware development.

Supply Chain Services

We offer a wide array of supply chain services including procurement, logistics, inventory management, temporary warehousing, programming, kitting and packaging services. We tailor our supply chain service offerings to meet the specific needs of our customers and enable our customers to manage supply chain planning and execution, which reduces costs and increases productivity. Our supply chain services are based on our proprietary software platform, which can be integrated with customers’ procurement management systems as well as our suppliers’ distribution management systems. Our global operating model enables us to support customers and their manufacturing ecosystems across key regions worldwide, providing responsive service wherever their products are built and deployed. Our global inventory management capabilities further enable us to manage a broad range of customer and supplier part numbers across manufacturing and logistics hubs worldwide, helping customers optimize inventory levels while maintaining reliable delivery and continuity of supply.

Optimized LED

Our Optimized LED segment offers a broad portfolio of application-optimized LEDs focused on improving lumen density, intensity, efficacy, optical control and/or reliability. Backed by expert design assistance and superior sales support, our Optimized LED products enable our customers to develop and market LED-based products for lighting, video displays and specialty lighting applications. This segment was formed after our acquisition of Cree LED in March 2021 and offers products under the Cree LED® brand. Our Optimized LED segment had net sales of $248.7 million, $256.1 million, and $259.8 million in 2026, 2025, and 2024, respectively.

Cree LED

For over 30 years, Cree LED has been a leader in LED lighting technology, offering innovative and differentiated LED solutions to a broad base of customers across multiple market segments. The Cree LED product offering includes both chips and packaged LED components.

Our Cree LED chip products include blue and green LED chips based on gallium nitride and related materials. These chips are used in a number of applications and are available in a range of brightness levels, wavelengths (colors) and sizes. Products using our blue and green LED chips can be found in automotive, video displays,

9

gaming displays, heart rate monitors and function indicator lights. Customers also combine our blue LED chips with phosphors to create white LEDs, which are used for indoor and outdoor illumination, medical and industrial applications and automotive interior lighting, indicators, headlamps and daytime running lights.

Our Cree LED XLamp®, J Series® and high-brightness product lines feature packaged LED components. The XLamp and J Series components meet a broad range of market needs for lighting applications, including general illumination (both indoor and outdoor applications), portable, architectural, signal and transportation lighting. Our high-brightness LED components consist of surface mount device (“SMD”) and through-hole packaged LED products. Our SMD LED component products are available in a full range of colors and are suitable for a variety of applications, including video, signage, general illumination, transportation, gaming and specialty lighting. Our through-hole packaged LED component products are available in a full range of colors and are primarily designed for the signage market.

Manufacturing and Test

We have manufacturing and testing facilities that support our business segments, which are primarily located in the United States, Malaysia and China. Our manufacturing facilities in Newark and Fremont, California and Penang, Malaysia, which support our Integrated Memory and Advanced Computing operations, are all certified in one or more of the following: ISO 9001:2015, ISO 14001:2015 and ISO 45001:2018. We also have a manufacturing facility in Huizhou, China for Optimized LED that is ISO 9001:2015, ISO 14001:2015 and IATF 16949:2016 certified. We are a member of the Responsible Business Alliance (“RBA”), and our manufacturing facilities in Malaysia and California currently comply with the RBA Code of Conduct, which increasingly is a business requirement of our customers.

Product testing is an important aspect of our manufacturing operations. We have established substantial technical expertise in the testing of products for high-end applications. Our extensive testing capabilities not only help to ensure a low defect rate but also enable us to sell specialized testing as an additional service. We design customer-specific testing processes that differ from the core focus of standard providers. We have achieved stringent quality targets across a broad spectrum of system applications and customer-specific designs. Our staff includes experienced test engineers who have developed proprietary testing routines and parameters that, when combined with our advanced test equipment, enable us to diagnose problems in components as well as system design, characterize the performance of new products, and provide high quality products at volume.

Advanced Computing

Within Advanced Computing, we utilize three primary methods of fulfilling product demand: using third-party contract manufacturers, building products to order based on sourced components, and configuring products to order. In each case, we use components and subassemblies acquired from a wide range of vendors. For our fault-tolerant products, we primarily use third-party manufacturers with assembly in our Ireland facility for certain product lines. For our AI and HPC products, we have developed capabilities for design and development of large-scale systems and dense AI and HPC clusters that have significant power and cooling requirements, with manufacturing and testing conducted at our Fremont, California facility. Our extensive experience in manufacturing and validating AI and HPC clusters has equipped us with the playbooks and best practices needed to move quickly.

Integrated Memory

Our Integrated Memory manufacturing operations benefit from our many years of design experience and our existing library of proven designs that stress high manufacturability and quality. Over 30 years of manufacturing experience enables us to move quickly to high-volume production of new products, which is paramount in helping our customers achieve rapid time-to-market for their latest innovations. Our design efficiencies, high level of automation, and expertise in advanced manufacturing processes power our build-to-order approach and help enable us to achieve high manufacturing yields, reduced direct labor costs, and quick turnaround of both small and large production orders.

For certain customers in our Integrated Memory segment, we employ extensive software-based electrical and thermal simulations and test our designs on high-end functional testers utilizing a broad array of test suites. These

10

tests are designed to meet the quality requirements of enterprise-class systems with stringent specifications required for various high-speed and high-compute applications. We also conduct design verification testing of hardware, firmware, system integration and reliability. We continually work to improve our test routines and associated software. For our specialty memory products, we have developed a high-volume, fully automated reliability testing and screening capability that substantially exceeds standard industry practices. These capabilities enable us to reduce the occurrence of early-life failures and weak module fallout, which can save our customers from the significant expenses associated with replacing products that fail after field deployment.

Optimized LED

While we do not own or operate wafer fabrication facilities, we have capabilities for subsequent stages of the product manufacturing cycle. In our manufacturing facilities in Huizhou, China, we receive LED chips from third-party wafer fabrication facilities, prepare and package dies into LED components, test components, and in some cases, assemble components on substrates or printed circuit boards to manufacture LED products. Through our years of investment and experience, we have developed expertise in LED technology that results in brighter, more efficient and lower cost LED chips and components for a broad range of lighting applications.

Customers

We believe that our customers look to us as a strategic partner based on our high-performance, high-availability, application-specific products, quality, technical support and global footprint. We also provide customized, integrated supply chain services for certain customers to assist them with the management and execution of their procurement and distribution processes. We believe our close collaboration with customers, customer-specific designs, long-lifecycle solutions and proprietary supply chain services create significant customer value. Our products are generally manufactured on a build-to-order basis. Our product sales are made primarily pursuant to customer purchase orders and are not based on long-term supply agreements. In our Advanced Computing segment, we also enter into multi-year managed services engagements, which typically span three to five years and include ongoing infrastructure management, optimization and support services.

We sell our products and solutions directly and through third-party channels, including distributors, value-added resellers, independent software vendors and systems integrators, to a diversified base of local and global OEM, enterprise and government customers.

In Advanced Computing, we sell our AI and HPC products to enterprise and government customers in AI cloud services, energy, financial services, hyperscale, and education; and our fault-tolerant computing products and services to enterprise customers in financial services, telecom, energy, transportation, healthcare, retail and industrial automation.

In our Integrated Memory segment, we sell memory solutions to OEM customers in industrial, government, networking and communications, enterprise storage and compute, and other vertical markets.

Our Optimized LED segment sells LED chips and components to manufacturers and electronic component distributors. We sell our Optimized LED products and solutions using both channel and direct sales to a diverse base of local and global OEM and contract manufacturers that together support a broad range of customers. We also utilize third-party sales representatives that generally do not maintain product inventory. A substantial portion of our Optimized LED products are sold to distributors that stock inventory and sell our products to their customers, which include value-added resellers, manufacturers who incorporate our products into their own manufactured goods and end-users of our products. Like other global semiconductor component suppliers, our Optimized LED segment has historically experienced seasonally lower sales during the earlier part of the calendar year and may see similar trends in the future.

In 2026, 2025, and 2024, sales to our ten largest end customers (including sales to contract manufacturers and original design manufacturers (“ODMs”) at the direction of such end customers) accounted for 56%, 66%, and 58% of total net sales, respectively. See “Item 1A. Risk Factors – Risks Related to Our Business – We depend on a select number of customers for a significant portion of our revenue” and “PART II – Item 8. Financial Statements and Supplementary Data – Notes to Consolidated Financial Statements – Concentrations.”

11

Suppliers

To address the needs of our customers, we have developed and maintained relationships with leading suppliers and contract manufacturers located in Asia, Europe and the Americas. Our semiconductor suppliers include many of the world’s largest memory manufacturers, including Samsung Semiconductor, Inc., Micron Technology, Inc., SK hynix, Inc., International Business Machines Corporation, and Kioxia Holdings Corporation. They also include some of the world’s largest providers of computing, communications and graphics processors, including Intel Corporation (“Intel”), Advanced Micro Devices, Inc. and TD SYNNEX Corporation; as well as providers of subsystems including Intel; networking products including Super Micro Computer, Inc. and Juniper Networks, Inc.; and suppliers of software products. Our contract manufacturers include NEC Corporation, Advantech Co., Ltd. and Celestica Inc. We often work with our suppliers in bidding for customers’ design-in opportunities. We also work closely with our suppliers to better ensure that needed materials are available and delivered on time. Our established global network of materials sourcing is designed to help ensure that our pricing remains competitive and to allow us to provide a stable source of supply for our customers.

We believe that our longstanding relationships with leading suppliers put us in a favorable position to procure sufficient quantities of materials, including during periods of industry shortages. Our flexible and responsive global manufacturing capabilities, inventory management systems and global IT systems allow us to move materials from one site to another in a cost-efficient manner and often deploy what might otherwise be excess inventory among other products and customers. In our Advanced Computing and Integrated Memory segments, we purchase a significant portion of our materials from suppliers on a purchase order basis and generally do not have long-term commitments from our suppliers. Our Optimized LED segment has a number of supply agreements with third-party providers for LED chips in wafer or discrete form.

See “Item 1A. Risk Factors – Risks Related to Our Business – We depend on a small number of sole or limited source suppliers.”

Global macroeconomic headwinds and evolving industry dynamics have contributed to supply constraints across our business segments. These supply challenges, combined with geopolitical factors including tariffs, have increased material costs and, in some cases, affected delivery timelines.

Sales, Support and Marketing

We sell our products both directly and through third-party channels to global OEMs as well as enterprise, government and other end customers located across North America, Asia, and Europe. Our sales and marketing efforts are conducted through an integrated process that engages our direct sales force, e-commerce, customer service representatives and on-site field application engineers (“FAEs”) alongside a network of independent sales representatives, distributors, integrators and resellers. Larger customers are also often supported by dedicated sales and support teams. Our sales and marketing efforts also include a high level of involvement from our senior executives.

Our on-site FAEs work closely with our sales team to provide product design support to our customers. Our FAEs collaborate with our customers, providing us with insight into their business models and product roadmaps and enabling us to identify early-stage opportunities that can help grow our business. Our integrated sales network and FAEs together allow us to be more responsive to our customers and more successful in navigating each customer’s unique and often complex design qualification and/or bid proposal processes.

Our marketing activities include advertising in technical journals; publishing articles in leading industry periodicals; hosting webinars; publishing white papers, electronic newsletters and blogs; and utilizing direct email solicitation. We participate in many industry trade shows worldwide and have active memberships in a number of industry organizations, including the Joint Electron Device Engineering Council, SD Card Association, Storage Networking Industry Association, CXL Consortium, Trusted Computing Group, Open Compute Project, Sensor Open Systems Architecture Consortium, Peripheral Component Interconnect Special Interest Group and Illuminating Engineering Society.

12

The timely development of new products and services is essential to maintaining our competitive position. Our primary research and development (“R&D”) activities are conducted at our R&D centers in the United States (Durham, North Carolina; Fremont, California; Irvine, California; Maynard, Massachusetts; Newark, California; Tempe, Arizona; and Tewksbury, Massachusetts) and in Huizhou, China; Bangalore, India; New Taipei City, Taiwan; and Penang, Malaysia. Our R&D activities focus on driving innovation in our products and services as well as continuous process improvement for procurement, test and manufacturing.

Our research and product development for Advanced Computing includes high-availability server architecture and design, high-availability software development (including virtualization, operating systems and systems management); server selection and occasional design; designs to enable integration of racks and clusters, including power and cooling design, evaluation, and deployment; storage system design and evaluation; high-performance network design; component testing for switches; cables and interface devices; development of software-defined storage systems; and embedded computer boards and systems.

Our product development in Integrated Memory includes innovations for next-generation DRAM products, including DDR5, DRAM, and emerging interconnect standards such as CXL, enterprise memory and flash-based products, as well as associated firmware development. We continue to develop a broad offering of flash-based products for industrial, government, communications and enterprise storage and compute markets. Our engineering team is focused on firmware development, systems engineering and integration, system and platform validation, applications, and product and reliability engineering for new products.

R&D in our Optimized LED segment includes innovations for next-generation LED products, including chips, packages for high-power general illumination, next generation direct-view video displays and specialty applications such as horticulture lighting, architectural, torch and emergency vehicles. We plan to continue R&D focused on the innovation and design of these and other new products that address the needs of our customers, with a focus on faster-growing markets.

Our advanced engineering and design capabilities enable us to address our customers’ increasingly complex needs. We design our products to be compatible with existing industry standards and, where appropriate, develop and promote new standards and provide custom solutions to meet customers’ requirements. An important aspect of our R&D is understanding the challenges presented by our customers’ requirements and addressing them using our industry knowledge, proprietary technologies and technical expertise. By working closely with our customers and suppliers, we are able to deliver technically advanced products designed to meet customer-specific needs with competitive solutions that satisfy our customers’ memory, storage and compute requirements; shorten their time-to-market; and enhance the performance of their end products and applications.

R&D expenses were $81.7 million, $79.8 million and $81.5 million in 2026, 2025 and 2024, respectively. As of August 28, 2026, we had approximately 380 R&D personnel worldwide.

Competition

Our businesses compete with numerous global and local companies. The principal competitive factors in our markets include the ability to meet customer-specific requirements and provide high product quality, strong technical support, technologically advanced products and services, advanced testing capabilities, flexible and global delivery options, reliable supply and reasonable pricing.

Across our Penguin Solutions business segments, our principal competitors include:

•Data center AI Infrastructure designers, integrators, and operators;

•Memory pooling products for AI inferencing;

•Compute and storage systems providers;

•Specialty memory products providers;

•Memory semiconductor manufacturers that also manufacture DRAM modules and Flash products;

•Supply chain service providers, including distributors and third-party logistics providers;

•Semiconductor and subsystem manufacturers;

•Embedded computing platforms and systems providers;

•Providers of System-on-Modules (SoMs) and Single-Board Computers (SBCs);

•LED product manufacturers;

13

•Enterprise IT server vendors; and

•Manufacturers of industrial computers.

Within Advanced Computing, we compete with global providers of AI infrastructure, compute systems, software, integration, and managed services, as well as manufacturers of enterprise servers and fault-tolerant computing solutions.

Within our Integrated Memory segment, we compete against memory module providers and, to a lesser extent, large semiconductor manufacturers that utilize a portion of their capacity to manufacture memory modules.

Within our Optimized LED segment, we compete with companies that manufacture and/or sell nitride-based LED chips as well as manufacturers of LED components concentrated primarily in indoor and outdoor lighting; specialty lighting including torch lamps (flashlights), horticulture and color-changing architectural lighting; signs and signals; and transportation.

We believe our differentiation comes from integrating advanced compute, memory, infrastructure software, and end-to-end services into a unified AI Factory Platform. This approach combines product innovation, deep technical expertise, and operational capabilities to help customers accelerate deployment, optimize performance, and manage AI infrastructure at scale.

In addition, some of our competitors are also our suppliers or customers. See “Item 1A. Risk Factors – Risks Related to Our Business – We depend on a select number of customers for a significant portion of our revenue” and “– We depend on a small number of sole or limited source suppliers.”

Intellectual Property

Intellectual property is an important aspect of our business. We actively seek to protect and leverage our intellectual property to promote our business interests. As of August 28, 2026, we owned or exclusively licensed approximately 1,619 patents, which are set to expire between 2026 and 2051, and had 531 pending patent applications. We continually review our development efforts to assess the existence and patentability of new intellectual property, which we work to protect as appropriate. We believe the duration and scope of our most relevant patents are sufficient to support our business, which as a whole is not significantly dependent on any particular patent or other intellectual property rights.

To protect our intellectual property, we rely upon a combination of patent, copyright, trade secret and trademark laws; contractual restrictions such as nondisclosure agreements, licenses and intellectual property assignment agreements; and policies and procedures. We pursue the registration of our domain names and trademarks in various jurisdictions and register trademarks in the United States and other countries as warranted. We maintain a policy requiring our employees, contractors, consultants and other third parties to enter into confidentiality and proprietary rights agreements to control access to our proprietary information. Additionally, we actively monitor data on our computer networks to assure compliance with data use policies. However, these laws, procedures and policies provide only limited protection and any of our intellectual property rights may be challenged, invalidated, infringed or misappropriated. Furthermore, the laws of certain countries do not protect proprietary rights to the same extent as the laws of the United States, which means that we may be unable to protect our proprietary technology in certain jurisdictions.

While many of our products contain proprietary aspects and are protected by patents, some of our products are built around mature industry standards and have less patent protection. For these products, we rely on trade secret rights to protect our proprietary interests. The absence of patent protection means that we cannot prevent our competitors from reverse-engineering and duplicating those products. Moreover, some of our product solutions incorporate open source software that is available under public licenses such as the GNU General Public License. We maintain policies and procedures to evaluate open source software used in our products and strive to minimize the risk of our proprietary intellectual property being inadvertently pulled under such licenses.

Human Capital

At Penguin Solutions, we aim to put people first by nurturing a culture that supports creativity and growth. We value our employees and understand the importance of their contributions as well as personal development. We believe that at the best workplaces, employees feel inspired, engaged, valued and included. As part of our “people

14

first” mindset, we are dedicated to advancing our people strategy by developing our talent, maintaining an engaged workforce and offering programs and rewards that attract and retain the best talent. At the core of our people strategy is a commitment to maintaining a safe and inclusive culture where individuals can achieve their highest potential, regardless of background.

Employees

As of August 28, 2026, we had approximately 2,800 full-time employees (excluding contractors) in locations across the globe, including in the United States, China and Malaysia. We have never experienced a work stoppage at any of our locations and take pride in our good employee relations.

Employee Engagement and Development

Continuous growth requires ongoing investment in people, innovation and new opportunities. We are always working to improve our communications between employees and management teams in order to advance our company goals and enhance the employee experience. We aim to develop capable leadership that can meet the challenges of business growth while instilling a supportive and inclusive company culture. At all locations, we provide our employees with performance assessments and evaluations. We place a priority on identifying talent throughout the organization and providing them with development and coaching opportunities through our newly adopted annual talent calibration and succession planning process. We also provide employees with access to technical and leadership training and training on workplace culture and enrichment, covering topics such as harassment, healthy work environments, inclusion, and global ethics and compliance.

Inclusion and Belonging

We strive to promote a workforce where individuals feel welcome and included. We know that a sense of belonging can help us to achieve our highest potential, helping to consistently raise the bar and drive innovation forward. This philosophy applies to all levels of our organization, including our executive leadership team and our Board of Directors.

Our focus on providing employees with an inclusive and nondiscriminatory work environment is outlined in our non-discrimination policy, though for the avoidance of doubt such policy is not incorporated herein by reference. Through this policy, we aim to articulate people-oriented, fair treatment principles for the recruitment, promotion, performance evaluation, compensation, training and retirement of all employees. We have established a council comprised of employees from across the organization focused on creating a working environment that offers equal opportunity employment so that all workers are treated with fairness and respect. The council enables us to expand on our strategy to nurture an inclusive work environment through initiatives such as employee resource groups which are open to individuals regardless of background, and efforts to recruit leadership from the widest pool of eligible talent.

Wellness, Health and Safety

We strive to provide and maintain a safe work environment and prioritize employee well-being throughout our operations. Our wellness program is supported by a committee that provides information and events throughout the year on health as well as mental and financial well-being. We also support mental health and wellness through our Employee Assistance Program, which offers free and confidential counseling and support for our employees and members of their households.

Compensation and Benefits

We offer compensation and benefits programs designed to motivate and reward our employees based on our pay-for-performance philosophy. We are committed to paying market competitive wages to attract key talent and work to help promote pay equality across equal quality and levels of work. Our bonus program links employee compensation to Penguin Solutions’ business performance. We also offer an Employee Stock Purchase Plan, equity compensation, retirement benefits and, in the United States, a 401(k) match program.

Environmental Regulations

15

Our operations and properties are subject to various federal, state, local, foreign and international environmental laws and regulations that govern, among other things, environmental licensing and registries, protection of flora and fauna, air and noise emissions, use of water resources, wastewater discharges, management and disposal of hazardous and non-hazardous materials and wastes, reverse logistics (take-back policy) and remediation of releases of hazardous materials. We cannot be certain that future identification of environmental concerns or conditions, more vigorous regulatory enforcement, enactment of more stringent laws and regulations or other unanticipated events will not arise in the future. Additionally, certain environmental laws may impose liability without regard to fault or the legality of the original conduct. These developments could give rise to material environmental liabilities and related costs that could have a material adverse effect on our business, financial condition and results of operations. See “Item 1A. Risk Factors – Risks Related to Our Business – We could incur substantial costs or liabilities as a result of violations of environmental laws.”

Available Information

Our address and principal U.S. executive offices are located at 45800 Northport Loop West, Fremont, California 94538 and our telephone number at this address is (510) 623-1231. Our principal website is www.penguinsolutions.com. Information contained on or accessible through our website is not a part of this Annual Report.

Through our website, we make available free of charge our Annual Reports on Form 10-K, Quarterly Reports on Form 10-Q and Current Reports on Form 8-K, and amendments to those reports, filed or furnished pursuant to Section 13(a) or Section 15(d) of the Exchange Act, as soon as reasonably practicable after they have been electronically filed with, or furnished to, the SEC.

The SEC maintains a website (www.sec.gov) containing reports, proxy and information statements and other information regarding issuers that file electronically with the SEC.

16