Envision announced on November 2, 2026, that it has commercially deployed an integrated 800-volt direct-current power system at its Galaxy AI data center campus in Ulanqab, China. The company described the installation as the first commercial deployment of its kind in a large-scale AI data center. It is supplying live computing workloads from grid input through to computing loads.
The architecture combines solid-state transformers, a battery energy storage system and equipment that adjusts DC voltage for computing hardware. Envision said its Dubhe Controller, supported by the Dubhe Energy Foundation Model, coordinates the transformers and batteries in real time to align power supply with computing demand and manage abrupt load changes.
The solid-state transformers convert power directly from 10 kV medium voltage to an 800V DC bus across operating conditions. Battery storage provides second-scale power support and minute-scale backup, while millisecond-level coordination between the storage system and transformers is intended to absorb GPU load fluctuations locally. According to Envision, the architecture can improve power conversion and watt-to-bit efficiency while reducing equipment footprint, cabling and copper requirements.
The deployment addresses the need to deliver more power to individual racks within limited space while maintaining stable supply as AI computing loads change. Envision said simplifying the power chain could lower infrastructure costs and allow greater computing density in the same area.
The Galaxy campus began operations in August 2026 and has a planned capacity of 3 GW. At full build-out, it is designed to support as many as one million AI accelerators operating in parallel. More than 80% of its power is planned to come from direct renewable electricity connections.
Envision plans to scale the approach through its Mission Gobi initiative, which targets 5 GW of green AI data center capacity in desert regions worldwide by 2030.
Source: prnewswire.com
