Wiwynn

What Wiwynn does

Wiwynn Corporation (TWSE: 6669) is a Taiwan-headquartered, publicly listed cloud IT infrastructure provider spun off from Wistron Corporation in 2016. The company designs, manufactures, and integrates rack-scale AI servers, storage systems, and data center hardware primarily for hyperscale cloud providers — Google Cloud, AWS, Meta, and Microsoft. Wiwynn is a tier-1 system builder in NVIDIA's Vera Rubin NVL72 ecosystem and AMD's Helios platform, and is one of the largest contributors to the Open Compute Project with 21 OCP Accepted certifications and 9 contributed designs.

The product portfolio spans 21-inch OCP-compliant servers (compute, storage, all-flash NVMe), 19-inch edge AI and 5G/vRAN servers, AI infrastructure platforms (NVIDIA Vera Rubin NVL72, AMD Helios/MI450X, SCADA accelerated storage server), and advanced thermal and interconnect technologies including two-phase immersion cooling (PUE <1.01), double-sided cold plate with liquid metal TIM, double-wide rack architecture, HVDC power delivery, and co-packaged optics integrated with Ayar Labs targeting >1,024 GPU scale and 100Tbps optical connectivity. The company holds 411 granted patents with 17.2% application growth in 2024, and a Universal Cooling Management System (UMS) provides intelligent monitoring across cooling solutions.

Wiwynn operates a direct enterprise sales motion selling fully integrated, pre-tested rack-scale systems on quote-based, long-term contracts to hyperscalers, with no reseller or distributor channel. Revenue is generated primarily from AI server hardware sales supplemented by rack integration and value-added services (thermal/power testing, burn-in, full-rack delivery). Global manufacturing spans Taiwan (HQ), United States (El Paso, $152M / 866K sq-ft campus), Mexico, Malaysia, and Czech Republic, enabling in-region delivery and nearshoring. Reported Q2 2025 revenue reached NT$220.74 billion (~$6.8B USD equivalent) at 185% YoY growth and Q3 2025 grew 172.8% YoY, with order visibility reported through 2027. A ~$2B USD convertible bond was launched in March 2026 to fund expansion.

Name

Wiwynn

Legal name

Wiwynn Corporation

Website

https://wiwynn.com

Company type

Public

Founded year

2016

Operating status

Operating

Headcount range

5,001–10,000 employees

Short description

Wiwynn is a Taiwan-listed ODM designing and manufacturing rack-scale AI servers, storage, and advanced cooling systems primarily for hyperscale cloud providers — Google, AWS, Meta, and Microsoft — across global facilities in Taiwan, the US, Mexico, Malaysia, and the Czech Republic.

Ownership category

akta.pro rank

Product category

AI Server Hardware

NAICS

Computer and Peripheral Equipment Manufacturing (33411), Computer Storage Device Manufacturing (334112), Ventilation, Heating, Air-Conditioning, and Commercial Refrigeration Equipment Manufacturing (33341)

SIC

Electronic Computers (3571), Air-Cond & Warm Air Heatg Equip & Comm & Indl Refrig Equip (3585)

akta.pro primary industry

Hyperscale & OCP Server Designs (HDABADAB)

akta.pro secondary industries

AI Datacenter Power, Cooling & Racks (liquid cooling, power delivery) (HDAAAAAH), Cooling Systems & Thermal Management (CRAC/CRAH, Chillers, Liquid Cooling) (HDABAKAD)

Keywords

• Rack-scale AI servers
• Data center cooling
• Hyperscale server hardware
• OCP-compliant infrastructure
• Liquid cooling systems

Headquarters

HQ city

Taipei

HQ country

Taiwan

HQ region

Asia

Offices6 records

Markets served

GTM type

B2B

Offering type

Hardware or Manufacturing

Cost components

Supply Chain, Technology or R&D, Personnel, Operations, Infrastructure, Marketing or Sales

Revenue model

1. AI Server Hardware Sales: Wiwynn generates the majority of its revenue from the sale of rack-scale AI servers, storage systems, and related rack infrastructure to hyperscale data center operators and cloud service providers. Products include liquid-cooled rack systems, OCP-compliant servers, and turnkey rack integration. The company reported record Q2 2025 revenue of NT$220.74 billion, up 185% YoY, driven by AI server demand.
2. Rack Integration and Value-Added Services: Wiwynn provides end-to-end rack integration services including thermal testing, power-draw efficiency testing, burn-in, quality and logistics support at the rack level. Full racks are tested as integrated units before deployment, providing TCO-optimized IT solutions. Services include proposal development, production, and in-region full rack delivery.

Go-to-market motion2 records

Distribution channels2 records

Marketing channels5 records

Core offering

Wiwynn designs, manufactures, and globally deploys rack-scale AI servers, OCP-compliant 21-inch and 19-inch cloud servers, GPU-accelerated storage systems, and advanced liquid cooling technologies for hyperscale cloud providers and large-scale data center operators. The company delivers end-to-end rack integration services—including thermal and power testing, burn-in, and full-rack shipment—covering the complete lifecycle from solution design to in-region deployment.

Product overview

Wiwynn is a cloud IT infrastructure provider offering a comprehensive portfolio of AI and data center solutions. The product lineup includes: (1) 21-inch OCP (Open Compute Project) servers comprising SV7100G4 and SV7220G3 (multi-node compute platforms), SV7000G4/ST7000G4 (high-density storage systems), and ST7200 (all-flash NVMe storage); (2) 19-inch servers including ES200G2 and ES100G2 (edge AI servers with NVIDIA GPUs), ES200 and EP100G2 (edge/5G vRAN platforms), SV310G4, SV305A, SV302A, and SV328R (cloud-native and ARM-based servers); (3) Accessories including OCP Network Mezzanine Cards and PCIe 3.0 Storage Cards; (4) Software & Services including Rack Integration Service for end-to-end deployment; (5) Advanced Cooling Technologies including Two-Phase Immersion Cooling, Cold Plate Liquid Cooling, and Universal Cooling Management System (UMS); and (6) AI Infrastructure products like the SCADA server and Double-Wide Rack Architecture. The portfolio is designed for hyperscale data centers with solutions spanning AI training/inference, edge computing, 5G/vRAN, storage, and cooling technologies.

Differentiator

Problem solved

Functional benefit

Products and services

• SV7100G4 OCP-certified multi-node 21-inch server based on Yosemite V3 specification for Open Rack V2 architecture; supports up to 12 blades in a 4OU chassis with Intel Xeon Scalable processors and enhanced AI inference via Intel Deep Learning Boost and bfloat16/AVX-512 support. Targeted at hyperscale cloud providers.
• SV7220G3 2U 3-node 21-inch OCP performance-optimized server for compute-intensive applications; supports Intel Xeon Scalable processors with Intel Deep Learning Boost (VNNI) for AI inference acceleration and up to 3TB DDR4 memory capacity.
• SV7000G4 / ST7000G4 High-density 21-inch storage system supporting up to 72 3.5-inch HDDs in a 4OU chassis; available in storage server and storage JBOD configurations using SAS-4 24Gbps IOCs for warm storage, cold storage, and big data applications.
• ST7200 2U OCP all-flash NVMe storage system supporting up to 30 NVMe SSDs, delivering up to 26.6 GB/s throughput and 5,984K IOPS (4KB random write); features tool-free modular design for easy maintenance and SSD upgrades.
• ES200G2 2U short-depth single-slot edge AI server with 5th Gen Intel Xeon processors and two NVIDIA L40S Tensor Core GPUs; optimized for AI inference/training and unified vRAN platforms; NEBS Level 3 compliant for carrier-grade deployments.
• ES100G2 1U short-depth single-slot edge server with 5th Gen Intel Xeon processors and two NVIDIA L4 Tensor Core GPUs; designed for edge AI inference and vRAN/NFV applications; NEBS Level 3 compliant with NVIDIA GPU Cloud readiness.
• ES200 2U short-chassis (450mm depth) edge computing server for telecom environments; supports 3rd Gen Intel Xeon Scalable processors with up to 270W TDP; NEBS-3 certified with AI acceleration via Deep Learning Boost.
• EP100G2 Multi-function 19-inch platform for 5G CU/DU/5GC/MEC and private 5G environments; supports 4th Gen Intel Xeon Scalable processors up to 225W TDP; 3U short-depth chassis with configurable 1U/2U compute sled modules and Intel QAT/AVX for 5G acceleration.
• SV310G4 Cloud-native 19-inch server featuring 3rd Gen Intel Xeon Scalable processors (up to 80 cores) with Deep Learning Boost for AI workloads; supports Intel SGX for security and is designed for power efficiency.
• SV305A Single-socket 19-inch server with AMD EPYC 7003 series processors; delivers up to 25% performance improvement over previous mid-range products; supports up to 200Gbps networking and OCP NIC 3.0.
• SV302A 1U dual-node high-density 19-inch computing server with AMD EPYC 7003 series processors (up to 64 cores per node, 128 total); modular design for flexible workload configuration including AI inference and real-time analytics.
• SV328R 19-inch ARM-based server featuring two Ampere Altra 64-bit ARM processors with up to 160 cores at 3.0 GHz; designed for high power efficiency with up to 24 U.2 NVMe SSDs and OCP NIC 3.0 support.
• Rack Integration Service End-to-end rack integration service covering four phases (Exploring requirements gathering, Proposal solution design, Production rack integration and manufacturing, and Delivery/Deployment commissioning); includes thermal testing, power-draw efficiency testing, and burn-in, with in-region full-rack delivery to data center operators.
• Two-Phase Immersion Cooling Advanced cooling technology achieving 90% cooling energy savings and PUE below 1.01; suitable for AI and machine learning workloads in hyperscale data centers, featuring intelligent tank management with safety protections; deployed in production at Microsoft's Quincy data center.
• Cold Plate Liquid Cooling Direct liquid cooling solution for cloud data centers offering improved cooling performance while maintaining existing infrastructure; features 6kW double-sided cold plate with liquid metal thermal interface material delivering 30%+ thermal efficiency gains, patent-pending blind-mate connectors, and optimized flow resistance design.
• Universal Cooling Management System (UMS) Open and modular cooling management system supporting liquid cooling rack management (UMS100) and immersion cooling tank management (UMS110); provides real-time monitoring, cooling energy optimization, and leak detection for data center operators.
• SCADA Server with GPU-Accelerated Storage One of the industry's first Nvidia SCADA (SCaled Accelerated Data Access) servers addressing storage bottlenecks in AI data center inference and training workloads; supports up to 96 liquid-cooled E3.S SSDs delivering 2.949 petabytes of storage capacity and 528 million 4K IOPS aggregated random read speed, using Nvidia Vera CPU and four RTX Pro 6000 Blackwell GPUs with PCIe 6.x switches.
• Double-Wide Rack Architecture Rack architecture that doubles the width of traditional racks to support larger, higher-power AI chips such as AMD Instinct MI400 Series GPUs; incorporates HVDC power delivery and liquid-cooled busbars for enhanced energy efficiency and thermal management in high-density AI data centers.

Quantifiable outcome

• Q2 2025 revenue of NT$220.74 billion, up 185% YoY
• +5 more outcomes

Named customers4 records

Segments2 records

Ideal customer profiles2 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

Integration12 records

AI capability6 records

Feature7 records

Partnerships

Six partnerships are on record, tiered core.

• Shinwa Controls Co., Ltd.coreStrategic or Co-development Partner · 3 June 2026Formal strategic collaboration to deliver advanced liquid cooling solutions for next-generation data centers, building on three years of joint engineering validation and product co-development. Combines Wiwynn's cloud server architecture with Shinwa's thermal control technologies including high-efficiency Direct Liquid Cooling systems and coolant distribution units, targeting cloud service providers with a turnkey cooling roadmap to lower TCO and improve PUE.
• NVIDIAcoreTechnology or Integration · 26 May 2026Wiwynn is a tier-1 system builder in NVIDIA's Vera Rubin ecosystem, developing Vera Rubin NVL72 liquid-cooled rack platforms featuring NVIDIA Rubin GPUs and Vera CPUs. Wiwynn showcased the SCADA server with GPU-accelerated storage at Computex 2026, one of the first such systems in the industry.
• AMDcoreTechnology or Integration · 26 May 2026Wiwynn builds AMD Helios rack-scale solutions based on OCP ORv3 specifications, featuring AMD Instinct MI450X GPUs and 6th Gen AMD EPYC CPUs. AMD announced $10B+ investment in Taiwan ecosystem with Wiwynn as a key ODM partner for multi-gigawatt deployments beginning H2 2026.
• Wistron CorporationcoreStrategic or Co-development Partner · 26 May 2026Wiwynn's parent company and co-development partner for rack-scale AI servers, including the NVIDIA GB200 NVL72 liquid-cooled system claiming 25x lower TCO compared to previous GPU generations, and Vera Rubin NVL72 platform development.
• Ayar LabscoreTechnology or Integration · 11 March 2026Strategic partnership to develop co-packaged optics (CPO) for rack-scale AI systems, enabling integration of Ayar Labs' optical engines and light sources into Wiwynn's rack-level architecture. The collaboration targets over 1,024 GPU scale and 100Tbps optical connectivity, addressing bandwidth and power limitations of copper interconnects in AI data centers.
• MicrosoftcoreTechnology or Integration · 18 October 2024Microsoft became the first cloud provider to deploy two-phase immersion cooling in production at its Quincy data center, developed in partnership with Wiwynn. This deployment demonstrates Wiwynn's immersion cooling technology achieving PUE < 1.01 in a production hyperscale environment.

Scale indicators8 records

Recent moves6 records

Expansion highlights6 records

Direct peers

• Quanta Computer: Taiwanese ODM and direct competitor in rack-scale AI servers for hyperscalers (Google, AWS, Meta, Microsoft). Both companies build NVIDIA and AMD platforms and compete head-to-head for hyperscaler rack-scale program awards.
• Hon Hai (Foxconn): World's largest ODM, competing with Wiwynn in AI server manufacturing for hyperscalers, including GB200 NVL72 and next-gen rack platforms. Operates massive global manufacturing footprint similar to Wiwynn's diversification strategy.
• Inventec: Taiwanese ODM with hyperscale server business supplying cloud providers. Competes with Wiwynn for AI server and rack integration contracts, particularly on NVIDIA-based platforms.
• Wistron Corporation: Wiwynn's parent company and co-development partner, also a direct ODM competitor in server and storage manufacturing for hyperscalers. The corporate relationship creates both collaboration and competitive tension across shared customer accounts.
• ZT Systems (an AMD company): US-based rack-scale AI server manufacturer acquired by AMD in 2024. Direct competitor in hyperscale AI systems, particularly on AMD Instinct platforms where Wiwynn is also a tier-1 builder.
• Pegatron: Taiwanese ODM with server business serving cloud and enterprise customers. Competes with Wiwynn in rack-scale AI infrastructure and OCP-compliant server designs.
• Supermicro: US-based AI and hyperscale server manufacturer offering rack-scale systems for cloud providers and enterprises. Competes with Wiwynn on NVIDIA-based AI servers and liquid cooling solutions.

Broad incumbents

• Lenovo Group (Infrastructure Solutions Group): Global technology vendor with hyperscale server business under its ISG division. Competes with Wiwynn on rack-scale AI systems, though Lenovo's portfolio spans far broader enterprise and PC segments.
• Dell Technologies (PowerEdge / Integrated Solutions): Major enterprise and hyperscale server vendor offering PowerEdge XE-series AI servers and integrated rack solutions. Overlaps with Wiwynn at hyperscalers but operates a much broader storage, PC, and services portfolio.

Emerging players

• Vertiv Holdings: Global provider of data center power and thermal management infrastructure, including liquid cooling solutions. Competes with Wiwynn on cooling technology integration and increasingly on full rack-level thermal solutions for AI deployments.

Market position

Strengths5 records

Weaknesses5 records

Competitive moat6 records

Key risks6 records

Key highlights7 records

Customer concentration

• Xx.com/WiwynnCorp
• YouTubeyoutube.com/@Wiwynn
• Facebookfacebook.com/WiwynnCorporation
• LinkedInlinkedin.com/company/wiwynn
• Instagraminstagram.com/wiwynncorporation

Compliance14 records

Revenue estimate

Valuation estimate

Number of profiles

Profiles6 records

Subsidiaries1 record

Funding overview

Funding rounds1 record

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments5 records

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Wiwynn