Envision Runs Live AI Workloads on Integrated 800V DC Power at China Data Center Campus

Envision, a global green technology company, says an integrated 800-volt direct current power system is now powering live computing workloads at its Galaxy AI data center campus in Ulanqab, China.

The company describes the installation as the world's first commercial deployment of such a system in a large-scale AI data center, delivering power from grid input through to computing loads. The announcement was issued from Ulanqab on Nov. 2, 2026.

Where the Deployment Sits

The system is installed at Envision's Galaxy AIDC Campus in Ulanqab, Inner Mongolia. According to the release, the campus began operations in August 2026. Its planned capacity is 3GW, and it is designed to support up to one million AI accelerators operating in parallel at full build-out. More than 80% of its power is planned to come from direct renewable electricity connections.

How the Architecture Works

Envision said the integrated system carries power from medium-voltage AC input to the server rack. It combines solid-state transformers, battery energy storage and equipment that adjusts direct-current voltage for computing hardware.

The company said this 800V DC architecture significantly simplifies AI data center power systems. The company attributes several advantages to the design: higher power conversion and watt-to-bit efficiency, a more compact equipment footprint, and reduced cabling and copper requirements.

Taken together, Envision said, these can lower infrastructure costs and support higher computing density within the same space. It added that the architecture supports higher-density AI racks, allowing more computing power in the same footprint.

The release sets the deployment against an operational challenge facing the industry. As demand for AI computing accelerates, Envision said, operators must deliver more power to each rack within constrained space while keeping supply stable as computing loads swing sharply from moment to moment.

Three Capabilities at the Core

Envision said the system rests on three critical capabilities. The first is stable conversion by the solid-state transformers directly from 10 kV medium voltage to an 800V DC bus across operating conditions. The second is battery storage that provides both second-scale power support and minute-scale backup. The third is millisecond-level coordination between the transformers and storage so that GPU load fluctuations are absorbed locally. The company said the system is already operating commercially ahead of wider industry adoption, and that it offers a reference point for the industry's transition to 800V DC.

Software Coordination of Power and Compute

Envision said its Dubhe Controller, supported by the Dubhe Energy Foundation Model, coordinates the transformers and batteries in real time. According to the company, the purpose is to match power supply with computing demand and to manage sudden load swings.

Kevin Huang, Senior Vice President of Envision Energy and President of Energy Storage and AIDC Power Systems, framed the deployment around coordination between energy sources and computing demand.

"AI data centers need power systems that coordinate renewable energy, energy storage, and the grid to meet computing demand," Huang said. "Our integrated 800V DC system is now powering live computing workloads, demonstrating an end-to-end architecture that connects renewable energy supply with computing demand. This is a critical step in turning power-compute coordination from a strategic vision into an operational reality."

Scaling Through Mission Gobi

Envision said it intends to scale this approach through its Mission Gobi initiative, which targets 5 GW of green AI data center capacity in desert regions worldwide by 2030. The company presents the Ulanqab system as a working example of an architecture toward which the wider industry is moving, rather than as a prototype.

In Envision's own description, the integrated system is delivering power to live AI workloads today, from grid input through to computing loads, with solid-state transformers, battery storage and AI-driven controls operating as a single coordinated system.