TLDR
- Nvidia partners with Foxconn to build a $1.4 billion supercomputing facility in Taiwan by mid-2026.
- The 27-megawatt data center will be Asia’s first deployment of Nvidia’s Blackwell GB300 chips.
- Foxconn’s new Visionbay.ai unit will operate the facility using a GPU-as-a-Service business model.
- Foxconn manufactures 1,000 Nvidia AI racks per week and plans to expand production capacity.
- Nvidia is working with Foxconn to integrate AI technology into factory manufacturing operations.
Nvidia revealed plans for a massive Taiwan data center project. The chipmaker is teaming up with manufacturing partner Foxconn.
The $1.4 billion supercomputing facility targets a mid-2026 opening. It will become Taiwan’s largest advanced GPU cluster.
The data center operates on 27 megawatts of power. It features Nvidia’s latest Blackwell GB300 chip technology.
The facility marks Asia’s first GB300 AI data center deployment. Foxconn created a dedicated unit called Visionbay.ai to manage operations.
Neo Yao leads Visionbay.ai as its CEO. The business unit specializes in AI supercomputing and cloud services.
Rental Model Transforms AI Access
Alexis Bjorlin spoke at Foxconn’s annual tech day event. The Nvidia vice president outlined a shift toward compute resource rental.
“Renting compute resources may offer a far better return on investment, enabling flexibility,” Bjorlin said. Companies can adjust their compute power based on business needs.
Bjorlin suggested that GPU technology evolves too quickly for facility ownership. Renting provides more economic sense than building individual data centers.
Visionbay.ai will offer GPU-as-a-Service to enterprise customers. The platform includes GPU leasing and Nvidia-native software solutions.
The service provides a cloud-based AI App Store for users. Enterprises gain access to training, fine-tuning and inference capabilities without major upfront costs.
The model removes traditional barriers to AI adoption. Companies can deploy AI technology using Nvidia’s accelerated computing architecture.
Production Capacity Expands
Foxconn serves as Nvidia’s main AI rack manufacturer. The company produces specialized server racks for AI workloads.
Chairman Young Liu reported production at 1,000 AI racks weekly. Foxconn expects to boost that number in 2026.
Liu told Reuters the company invests $2 billion to $3 billion yearly in AI. Foxconn projects AI demand will drive major growth next year.
Spencer Huang attended Foxconn’s tech day alongside other industry leaders. He manages Nvidia’s robotics product line.
Huang discussed plans to bring AI into factory settings. Nvidia and Foxconn are collaborating on manufacturing line integration.
The tech day featured representatives from OpenAI and Uber. Foxconn founder Terry Gou also made an appearance.
Foxconn announced a separate OpenAI partnership at the event. The deal involves producing components for U.S.-based data centers.
The company has diversified beyond iPhone assembly work. Foxconn now develops electric vehicles and AI infrastructure.
The Taiwan facility advances sovereign AI infrastructure goals. The GB300 NVL72 systems deliver enterprise-grade computing at scale.
Bjorlin emphasized flexibility as GPU technology accelerates rapidly. The rental approach lets companies match spending to actual usage patterns.
Liu said Foxconn’s EV production reaches outsourcing viability levels. The company showcased its Model A electric vehicle at the tech day.
Get 3 Free Stock Ebooks
Discover top-performing stocks in AI, Crypto, and Technology with expert analysis.
- Top 10 AI Stocks - Leading AI companies
- Top 10 Crypto Stocks - Blockchain leaders
- Top 10 Tech Stocks - Tech giants
