Wiwynn
Wistron's hyperscale spinoff, focused purely on cloud datacenter systems. Wiwynn designs rack-scale liquid-cooled NVL systems for hyperscalers and has been moving up the stack into full rack integration. Not a channel vendor: it sells at datacenter scale.
- ODM-direct model: end-to-end from board-level components through complete L11 rack integration; OCP Platinum member
- One of four ODMs (with Foxconn, Quanta, Wistron) holding NVIDIA Certified Systems status as main suppliers of GB200/GB300 full cabinets
- Among the first in line for GB300 NVL72 and Vera Rubin NVL72 readiness, with fully liquid-cooled fan-less rack designs
- Own rack cooling engineering: Aqualoop ORv3 liquid cooling with blind-mate tubes, UMS100 cooling management integrated with Auras CDUs
| Model | Platform | Form factor | Cooling |
|---|---|---|---|
| SV7000G2 / SV7000G4 Up to 8 double-width GPUs over PCIe Gen5 | HGX B200 / HGX B300 | 4U | air or DLC |
| 10U HGX B300 system Blackwell Ultra | HGX B300 NVL16 | 10U | DLC |
| GB200/GB300 NVL72 cabinets Main supplier status alongside Foxconn, Quanta, Wistron | GB200 NVL72 / GB300 NVL72 | full rack | DLC |
| Aqualoop OCP ORv3 liquid cooling with blind-mate tube features | GB300 NVL72 | rack cooling infrastructure | DLC |
| ES200 OCP-inspired edge server | MGX | 2U short-depth edge | air |
- Showcased Vera Rubin NVL72, HGX Rubin NVL8 (fanless 2U, 100% liquid cooled) and RTX PRO Server at GTC 2026
- Among the first in line for GB300 NVL72 readiness (GTC 2025)
- Posted the highest revenue growth among top-20 ODMs on GB200 shipments and custom-ASIC AI servers
- Tainan, Taiwan (manufacturing center)
- Senai, Johor, Malaysia (rack integration plus PCBA plant)
- Brno, Czech Republic (Europe)
- Juarez, Mexico (North America)
- Socorro (El Paso), Texas: first US facility, $300M approved, 866,000+ sq ft, operations from H1 2026
- Anchor customers: Meta (reportedly over 50% of revenue) and Microsoft, both named
- ODM-direct to hyperscalers and CSPs, not enterprise channel
- FY2025 capex tripled to NT$18B; Q4 2025 gross margin 7.2% on product mix, a pricing signal for volume negotiations
Is Wiwynn an OEM, ODM or integrator for NVIDIA AI servers?
Wiwynn sits in the "ODM system designers" tier of our index. ODM-direct model: end-to-end from board-level components through complete L11 rack integration; OCP Platinum member
Does Wiwynn design its NVIDIA servers in-house?
Wistron's hyperscale spinoff, focused purely on cloud datacenter systems. Wiwynn designs rack-scale liquid-cooled NVL systems for hyperscalers and has been moving up the stack into full rack integration. Not a channel vendor: it sells at datacenter scale.
Which NVIDIA platforms does Wiwynn ship?
Wiwynn ships systems on HGX B200 / HGX B300, HGX B300 NVL16, GB200 NVL72 / GB300 NVL72, GB300 NVL72, MGX. The systems table on this page lists the specific models.
Where does Wiwynn manufacture its AI servers?
Tainan, Taiwan (manufacturing center). Senai, Johor, Malaysia (rack integration plus PCBA plant). Brno, Czech Republic (Europe). Juarez, Mexico (North America). Socorro (El Paso), Texas: first US facility, $300M approved, 866,000+ sq ft, operations from H1 2026