Wiwynn

San Jose – October 13, 2026 -- Wiwynn, Ayar Labs and Global Unichip Corp. (GUC) are showing the first live demonstration of a compute system that integrates GUC's ASIC with Ayar Labs' TeraPHY optical engine and SuperNova light source while streaming real-time health telemetry, at OCP Global Summit 2026.

Demo moves CPO from rack-level architecture to platform integration and functional validation

The demonstration advances the companies' earlier rack-level architecture collaboration. Wiwynn's system integrates GUC's multi-chip packages (MCPs) with TeraPHY optical engines (OEs), powered by the Ayar Labs SuperNova light source. Telemetry is continuously polled from the OEs inside the MCPs and displayed in real time, showing the optical link is up within the system.

San Jose – October 12, 2026 -- Wiwynn unveiled a 6kW double-sided cold plate that it says cuts power delivery loss by up to 80%, alongside AI rack platforms for NVIDIA and AMD and a co-packaged optics (CPO) rack, at the 2026 OCP Global Summit in San Jose.

Wiwynn joins NVIDIA Vera Rubin NVL72 pilot build program

Wiwynn and Wistron are among the first NVIDIA Pilot Build Partners for the Vera Rubin NVL72 platform, covering engineering validation and design-in. The liquid-cooled, rack-scale system integrates 72 NVIDIA Rubin GPUs and 36 NVIDIA Vera CPUs with 6th-generation NVIDIA NVLink.

Wiwynn is also applying its rack integration and manufacturing capabilities to AMD's Helios rackscale solution, built on the Open Rack Wide (ORW) standard. Helios combines AMD Instinct MI450 Series GPUs, 6th Gen AMD EPYC CPUs, AMD Pensando networking and AMD ROCm software, with configurations tailored to cloud service provider (CSP) requirements.

San Jose – October 12, 2026 -- Wistron Corporation and Wiwynn Corporation say they are among the first manufacturers ready to build NVIDIA's Vera Rubin NVL72 rack-scale platform, which unifies 72 Rubin GPUs and 36 Vera CPUs in a fully liquid-cooled design. The two companies are showing the system at the 2026 OCP Global Summit in San Jose under the theme "Create AI Beyond Limits."

Vera Rubin NVL72 combines 72 GPUs, 36 CPUs and BlueField-4 DPUs

The NVL72 platform links 72 NVIDIA Rubin GPUs and 36 NVIDIA Vera CPUs through NVIDIA NVLink. It also integrates NVIDIA ConnectX-9 SuperNICs and NVIDIA BlueField-4 DPUs. Wistron and Wiwynn describe it as built for the resiliency, energy efficiency and scalability that AI factories demand, and optimized for frontier AI model training, inference and reasoning.