The Fort Worth plant provides approximately 324,000 square feet of manufacturing space, equivalent to about 30,000 square meters, configured for high-volume assembly, integration, and testing of advanced Nvidia AI systems. Wistron positions the site as a cornerstone of Nvidia’s U.S. AI supercomputer manufacturing network in Texas, dedicated to board-level production and server rack integration for data center deployments.
The facility is designed as an AI “smart factory,” using Nvidia’s accelerated computing and digital twin technologies to optimize workflow, quality control, and throughput in superchip module production. This strategic move comes amidst a growing AI boom impact on the tech industry, highlighting the increasing demand for advanced processing capabilities.
Current production at the plant centers on Nvidia’s GB300 Grace Blackwell Ultra superchip modules, which combine CPU and GPU capabilities to power large-scale AI training and inference workloads. Wistron has already produced the first GB300 compute board manufactured in the United States, and the factory has entered mass production for these modules.
GB300 Grace Blackwell Ultra superchips drive large-scale AI training and inference in Fort Worth mass production
The roadmap for the Fort Worth site includes manufacturing Nvidia’s next-generation Vera Rubin Superchip in a dedicated production cell, extending the facility’s role in future AI computing platforms. These superchips are intended for deployment across hyperscale data centers, cloud service providers, and enterprise AI infrastructure throughout the U.S. market.
Nvidia and Wistron expect the Texas facility to ramp output over 2026 to tens of thousands of computing boards per month, classifying the site as a mass-production hub within Nvidia’s U.S. AI supercomputer roadmap. Production is targeted to support rapid deployment of AI server clusters, with systems integrated into Nvidia’s data center platforms for customers building large-scale AI factories.
The plant is embedded in a broader U.S.-based supply chain that links Nvidia design work in California, advanced chip fabrication and packaging in Arizona, and system assembly and testing in Texas. Partners including TSMC, Foxconn, Wistron, Amkor, and SPIL participate in this network, which is intended to deliver as much as $500 billion of AI infrastructure in the U.S. over four years.
The Fort Worth factory has created hundreds of manufacturing jobs and is expected to employ around 1,000 people as production scales. Across Nvidia’s U.S. initiative, its AI chip and supercomputer factories are projected to support hundreds of thousands of jobs nationwide. Together with additional Arizona and Texas sites dedicated to Blackwell chips and AI supercomputers, the new plant contributes more than one million square feet of U.S. AI manufacturing capacity.
