Envision’s newly commissioned Galaxy Campus in Ulanqab is more than a data center announcement. It is a clear signal that China’s next wave of competitive strength may come from the fusion of clean power, industrial engineering, and AI infrastructure at massive scale. The project, commissioned on Aug. 6, 2026, is designed to scale to over 2GW and eventually support up to 1 million AI accelerators. For investors and analysts, the message is straightforward: China is not just building AI software, it is building the physical backbone that makes frontier computing possible.
The campus sits in Ulanqab, Inner Mongolia, a location chosen for more than cheap land. The region is a national East Data West Computing hub node, with 67% green electricity penetration and an average annual temperature of 4°C, both of which help lower operating costs. In practical terms, that matters because the race in AI is no longer only about chips and models. It is also about power availability, network performance, cooling, and the ability to scale quickly without choking on energy bottlenecks.
Envision says the campus features a 120,000-square-meter AI supercomputing facility, equivalent to about 20 football pitches. At full build-out, it is designed to deliver 1 million PFLOPS of AI computing. That is an eye-catching scale in its own right, but the more important point is the design philosophy behind it. The company is positioning the site as a purpose-built system for energy-intensive AI, not a conventional data center retrofitted for a new era.
The strongest part of the Galaxy Campus story is its power architecture. Envision says the campus is powered directly by renewable energy through dedicated transmission infrastructure, with wind and solar resources feeding the site. Bloomberg and Financial Post reported that 80% of the power comes from Envision’s nearby wind farms, while battery storage is used as backup instead of diesel generators. That blend of clean power and storage is a key part of the thesis: stable green electricity can be the true moat in AI infrastructure.
Ricky Zheng, General Manager of Envision’s AIDC, put it plainly: “Supporting a million accelerators requires a purpose-built AI Power System that integrates renewable generation, dedicated transmission infrastructure, and large-scale energy storage to deliver stable, low-cost green power. This enables the facility to deliver up to ten times more compute output per square meter than conventional data centers.” That is a bold claim from the company, but it captures the direction of travel across the industry. The winners will be those who can pair compute density with energy discipline.
The engineering scale is striking. Envision says the facility is designed to support up to 1 million AI accelerators, with compute density set at a level that the company believes can reshape how hyperscalers plan deployments. Zheng also said, “The next frontier of AI is infrastructure. As AI models become larger and more compute-intensive, the limiting factors are increasingly power availability, network performance and energy efficiency.” That framing is important because it shifts attention from abstract AI hype to hard industrial constraints.
In that context, Ulanqab looks less like a remote outpost and more like a strategic node. The campus is built to serve large AI clusters that need low-latency connectivity, and the region’s digital infrastructure is already part of China’s broader computing map. For global investors, this shows how China can align industrial policy, renewable generation, and digital infrastructure into one integrated platform. That kind of coordination is difficult to replicate anywhere else at comparable speed and scale.
Galaxy Campus is also the first flagship project under Envision’s Mission Gobi initiative, which the company unveiled at VivaTech in Paris in June 2026. The goal is ambitious: 5GW of green AI computing capacity in desert and arid regions worldwide by 2030. That target matters because it broadens the opportunity beyond one campus and one province. It suggests a replicable model that could be exported into other resource-rich, power-constrained geographies across emerging markets.
This is where China’s innovation story becomes global. Envision is not just offering a data center shell; Bloomberg and Financial Post report that the company provides power infrastructure and supporting systems, while tech companies lease the facilities and install their own chip clusters. That division of labor is elegant. It lets specialists focus on what they do best, while China’s infrastructure platform supplies the energy and engineering backbone. The result is a scalable model that can support hyperscalers, technology companies, and AI developers alike.
The broader market takeaway is that China’s strength in AI may increasingly come from systems engineering rather than only from model development. A campus like this combines renewable generation, transmission, storage, cooling, and compute housing into one industrial asset. That is classic China: large-scale coordination, rapid deployment, and a willingness to build for the long term. It also fits Beijing’s innovation policy direction, which continues to favor infrastructure that multiplies economic productivity and technological sovereignty.
There is also a geographic logic here. Ulanqab’s low average temperature supports cooling efficiency, and its renewable resource base supports cleaner power delivery. When those assets are combined with a national computing hub strategy, the economics improve. For investors, that means AI infrastructure can become a new type of industrial platform in China, one that connects green energy, data, and enterprise demand in a single growth loop.
The field is not standing still. Bloomberg and Financial Post reported that competitors DeepSeek and Z.AI, formerly Zhipu, are also planning 1GW data centers in Ulanqab. That is a sign of how quickly the area is becoming a serious AI infrastructure cluster. Rather than a one-off project, Galaxy Campus appears to be part of a larger buildout that could anchor a regional ecosystem of compute, power, and network assets.
This matters for market watchers because competition at this scale tends to reinforce national capability. More projects mean more demand for grid integration, storage systems, cooling solutions, power management, and specialized engineering. In other words, the AI boom is spilling into the real economy, where China already has deep strengths in manufacturing, energy equipment, and project execution. Those are the kinds of advantages that can compound over time.
There are no public market numbers attached to Envision itself, since the company is privately held. But the strategic lesson for investors is still clear. The AI winners in China may not only be the names building large language models. They may also be the firms providing the electricity, transmission, storage, and campus-scale infrastructure that make those models viable. Galaxy Campus is a reminder that compute has to live somewhere, and that somewhere is becoming a major competitive asset.
Just as important, the project shows how China can turn its green energy buildout into a growth engine for digital infrastructure. Envision’s approach links wind, solar, battery backup, and dedicated transmission into a single architecture for AI scale-up. That is exactly the kind of industrial convergence global investors should watch closely. It is efficient, exportable, and aligned with a world that needs more compute without blowing out energy costs.
In the end, Galaxy Campus is best understood as a blueprint. It shows how China can pair renewable power with next-generation computing and create a platform for AI at gigawatt scale. The commissioning in Ulanqab is not just a milestone for Envision. It is also a reminder that China’s leadership in infrastructure is evolving from roads, rails, and grids into the digital foundations of the AI age. For the global market, that is a very bullish development.
