WhiteFiber, Inc. (NASDAQ: WYFI) announced the commercial availability of WhiteFiber Continuum on Sept. 23, 2026, a cross-data-center networking product the company claims is the first commercially available distributed GPU supercluster architecture.

The system links two HITRUST-certified QTS data centers located 83 kilometers apart using 12 Zayo dark fiber strands. According to the company, the architecture delivers 136 Tbps of bandwidth at 0.9 milliseconds of guaranteed round-trip latency. WhiteFiber has submitted patent applications for the underlying implementation.

The product is built in collaboration with DriveNets, which supplies the AI network fabric, and WEKA, which provides data storage and memory infrastructure across the cluster under its NeuralMesh platform.

"WhiteFiber Continuum is the infrastructure answer to a problem the industry has lived with for years: geographic distance as a ceiling on what a cluster can do," said Sam Tabar, Chief Executive Officer of WhiteFiber.

The company states the architecture is designed to allow enterprises to pool GPU resources across geographically separated sites, support workloads that require data sovereignty within a specific jurisdiction, and maintain training runs if one location goes offline.

WhiteFiber describes the product as suited for regulated industries and for operators of telecom and metro facilities that hold unused power and fiber capacity. The company also states it is designed to support last-mile inference and agent workloads.

The information in this article is based on a press release issued by WhiteFiber, Inc.