Wiwynn

緯穎 / 緯穎科技服務股份有限公司

A Taiwan-based original design manufacturer that builds and integrates the accelerated servers and high-density racks that hyperscalers use to deploy AI capacity.

TWSE: 6669
Profile as of September 13, 2026
#11 in the AsiaAI Tech Index
How the ranking works →

Wiwynn is a direct beneficiary of hyperscaler AI spending, converting accelerator allocations into deployable data-center capacity at global scale. Its central tension is executing high-volume, qualified system integration at low operating margins while remaining dependent on customer platform decisions and allocated silicon.

Key figures

Revenue FY2025: NT$950.663 billion
Revenue growth FY2025: 163.7%
Operating margin FY2025: 6.7%
Net income FY2025: NT$51.118 billion
R&D spend: Not disclosed
Capital expenditure: Not disclosed
Market capitalization 2026-09-11: NT$1.28 trillion Estimate
AI-related products share of revenue FY2025: More than 50%
First-half revenue H1 2026: NT$554.660 billion

Overview

Wiwynn Corporation, founded in 2012 and spun out of Wistron Corporation, is a Taiwan-listed original design manufacturer and systems integrator. It focuses exclusively on cloud service providers and hyperscale data centers rather than retail or broad enterprise hardware channels.

The company supplies rack-scale infrastructure, including GPU and ASIC server platforms, full-rack integration, high-power distribution, thermal architectures, and liquid-cooling solutions. It operates manufacturing facilities in Taiwan, Mexico, Malaysia, and Texas to support global cloud deployments.

Wiwynn's differentiation lies in system integration for hyperscale customers rather than proprietary semiconductor technology. It combines board-level design, server design, manufacturing engineering, and customer-specific qualification to deliver high-density data-center capacity.

The AI angle

Wiwynn is a direct AI-infrastructure beneficiary. As cloud service providers expand their AI computing clusters, demand increases for accelerated servers, rack integration, power delivery, and thermal management. The company states that AI-related products contributed more than half of its FY2025 revenue, driving a 163.7% year-over-year revenue increase to NT$950.663 billion.

The company does not design AI accelerators, own cloud capacity, or sell foundation models. Instead, it converts customer-selected accelerators—such as GPUs or ASICs—into deployable, qualified rack-scale systems. This requires synchronizing accelerator allocation, memory, networking, racks, cooling loops, factory testing, and localized manufacturing to meet hyperscalers' high-density power and liquid-cooling requirements. By bridging the gap between semiconductor suppliers and cloud operators, Wiwynn captures value from the AI build-out without taking on the risk of developing proprietary silicon or training frontier models. Its position in the AI Server Hardware and Power & Infrastructure layers makes it a critical enabler for hyperscale data centers that require specialized engineering to handle the extreme thermal and power demands of modern AI workloads.

Technology and moat

Wiwynn's moat is its qualified operating model and system-integration capability rather than proprietary semiconductor IP. It designs and integrates GPU and ASIC servers, board-level systems, racks, power delivery, high-speed signal paths, thermal systems, and liquid cooling for high-density cloud data centers. The company supports multiple accelerator architectures and works closely with technology partners across the networking and power ecosystems.

The difficult-to-copy element is the execution of high-volume AI deployments. Moving a hyperscale AI cluster from design to operation requires synchronizing component allocation, factory testing, logistics, firmware, validation, and localized manufacturing. Wiwynn's global production footprint across Taiwan, Malaysia, Mexico, and Texas reduces the operational friction of that transfer and provides resilience for its cloud customers. Replicating this requires years of customer-specific qualification and billions in supply-chain coordination.

Five-pillar assessment

Scale and market position: A major original design manufacturer for hyperscale data centers, delivering nearly NT$1 trillion in annual revenue with a global manufacturing footprint.
Technology and R&D: Specializes in system architecture, rack integration, power delivery, and liquid cooling for high-density AI servers, rather than proprietary semiconductor technology.
Supply-chain centrality: Deeply integrated into hyperscaler supply chains and Wistron's manufacturing ecosystem, though dependent on customer-selected silicon and accelerator allocations.
Financial momentum: Exceptional FY2025 revenue growth of 163.7% driven by AI infrastructure demand, alongside ongoing global manufacturing expansion.
Governance and quality: Publicly listed with independent directors, operating under the effective control of its parent company, Wistron Corporation.

Supply chain and relationships

Wiwynn depends on a supply chain of customer-selected accelerators, memory, networking silicon, power systems, and cooling equipment. It does not control the allocation of these high-value components, making it reliant on sufficient supply of GPUs and ASICs to meet its server shipment targets. It also relies on its parent company, Wistron Corporation, and its broader manufacturing ecosystem for production support.

On the customer side, Wiwynn's revenue is highly concentrated among global cloud service providers and hyperscale data centers. While specific customer names are not disclosed in its annual report, this market structure typically produces large-account concentration. The company acts as an intermediary, taking allocated silicon from companies like NVIDIA and delivering finished, rack-scale infrastructure to the world's largest cloud operators.

Suppliers

WistronParent company and manufacturing ecosystem partner

Partners

NVIDIAInferredAccelerator ecosystem participant for GPU servers

Competitors

Quanta Cloud TechnologyInferredCompetes for hyperscale cloud and AI server programs
FoxconnInferredCompetes in AI-server manufacturing and rack-scale systems
InventecInferredCompetes in cloud-server and AI-server system integration
Super Micro ComputerInferredCompetes in accelerated-server and rack-scale AI infrastructure
WistronInferredCompetes for server-system and AI-rack programs
MiTACInferredCompetes for server-system and AI-rack programs
CompalInferredCompetes for server-system and AI-rack programs
Hewlett Packard EnterpriseInferredHas relevant server and rack-integration capabilities

Geopolitics and risk

Wiwynn's most significant geopolitical exposure is its reliance on Taiwan for engineering and manufacturing, making it vulnerable to cross-strait disruption that could affect design, component flows, factories, logistics, and customer supply assurance.

The company is also exposed to U.S. export controls on advanced AI chips, which can constrain configurations and customer deployments in certain markets, particularly where controlled accelerators are central to the system bill of materials. To mitigate geopolitical and global trade risks, Wiwynn is actively expanding its production footprint outside of Taiwan. It operates facilities in Mexico and Malaysia, and a new plant in Texas began production in late 2025, providing North American customers with localized manufacturing and reducing single-region dependency.

Risk matrix
Risk	Severity	Why it matters
AI accelerator allocation	High	Shipments depend on sufficient supply of customer-selected GPUs and ASICs.
U.S. export controls	High	Restrictions on advanced AI chips can constrain configurations and deployments.
Cross-strait disruption	High	Taiwan remains important for engineering and manufacturing footprint.
Customer concentration	High	Serves a consolidated market of hyperscale data centers.
Margin compression	High	Low operating margins leave limited room for component-cost inflation or pricing pressure.
Global trade and tariff changes	Medium	Trade-policy shifts necessitate expanding production in multiple countries.
Manufacturing ramp execution	Medium	New Texas facility must secure customer qualifications and sustainable utilization.
Technology-transition risk	Medium	Rapid changes in accelerated computing architectures require continuous engineering investment.

Governance and ownership

Wiwynn is an independent publicly listed company on the Taiwan Stock Exchange. It operates under the effective control of Wistron Corporation, which holds a 35.46% equity stake and appoints directors to the board. The company is not state-owned, and there is no evidence of state influence in its operations.

It is led by Chairman and Chief Strategy Officer Emily Hong and President and CEO William Lin. The board includes executive and Wistron-affiliated directors, as well as five independent directors. Wiwynn reports under Taiwan exchange rules and maintains public financial reporting, with substantial institutional ownership alongside Wistron's controlling block.

What to watch

Whether AI-related products remain above half of revenue as growth normalizes from FY2025's 163.7%.
Whether operating margins remain near or above 6.7% amid changing component costs and AI-rack mix.
Whether the new Texas facility secures customer qualifications and economically sustainable utilization.
Whether the company can expand AI-server output without bottlenecks in accelerators, memory, and cooling components.
Whether the authorized capital raising of up to 18 million common shares occurs and affects share count.
Whether U.S. trade policy changes redirect customer production among its global sites.

Recent developments

August 7, 2026Reported Q2 2026 consolidated revenue of NT$278.153 billion and first-half 2026 revenue of NT$554.660 billion.
February 26, 2026The board approved FY2025 financial statements, reporting revenue of NT$950.663 billion.
December 31, 2025The Texas plant began production capacity by the end of Q4 2025.
August 8, 2025Approved a NT$4 billion headquarters construction budget and a US$500 million capital increase for Wiwynn International Corporation.
May 8, 2025William Lin was promoted to President and Chief Executive Officer.

Coverage on AsiaAI.FYI

We haven't written about Wiwynn yet. Subscribe below to get it when we do.

Guides that cover Wiwynn

About this profile

Compiled with AI-assisted research from company filings, market data, and published reporting as of September 13, 2026, then reviewed by AsiaAI.FYI. Figures marked Estimate are not company-reported. Check primary filings before relying on any number.

Confidence: B. Financial performance, leadership, and AI-revenue exposure are supported by company disclosures, but named end customers, supplier relationships, and exact AI-server market share are not publicly quantified.

Main sources: FY2025 annual report; FY2025 financial-results release; Quarterly financial releases; Shareholder-services disclosures; Market-data pages.

All 35 sources

Position in the AI stack

Integrates customer-selected GPUs, ASICs, memory, and networking silicon into qualified server systems and full cabinets for cloud service providers.

Legal name: Wiwynn Corporation
Headquarters: New Taipei City, Taiwan
Founded: 2012
Leadership: Emily Hong / 洪麗甯, Chairman and Chief Strategy Officer
William Lin / 林建勳, President and Chief Executive Officer
Employees: 9,138 (2024)
Ownership: Publicly listed; part of Wistron Corporation
Listings: TWSE: 6669
Business: Original design manufacturer and systems integrator for cloud and hyperscale data-center infrastructure
Sector: Servers, PCs & ODMs
Website: wiwynn.com