Published: June 04, 2026, 14:18 UTC
Chinese cities are accelerating their push into space-based computing at an institutional level. Beijing, Tianjin and Chengdu have each established dedicated research institutes and innovation consortia aimed at building what industry observers call the next frontier of AI infrastructure.
The moves signal that orbital computing is shifting from speculative startup pitches to formal national strategy. China is creating the institutional scaffolding to compete in what analysts predict will be a trillion-dollar global market by 2030.
Three Cities, Three Approaches
Beijing established its first space computing innovation center on June 1. The center adopts a company-plus-alliance model focused on six areas across the space computing value chain: chips, hardware, platforms, AI, networking, and applications. The goal is to create a coordinated industrial system that spans the full stack.
On the same day, the privately-owned satellite company GalaxySpace set up a dedicated space computing research institute in Beijing’s E-Town development zone. The institute targets advancements in space-borne computing chips, inter-satellite laser communications, space-based energy and thermal management, integrated space-ground networking, and space security standards. Its first experimental satellite is expected to launch by 2028.
Tianjin followed on May 29 at the World Intelligence Expo 2026. The National Supercomputing Center in Tianjin joined with companies and research institutes to form a joint consortium for space digital intelligence infrastructure. The consortium will focus on modular and scalable computing payloads, high-performance on-board chips, in-orbit intelligent operations, space computing software stacks, integrated energy-thermal control systems, and flexible space solar arrays.
Chengdu entered the race earlier, on March 26. The city’s aerospace company ADAspace, in collaboration with the Ministry of Industry and Information Technology (MIIT), launched Prometheus — the world’s first space-based computing power cloud service platform for enterprises.
The Big Picture
China’s orbital computing plans extend well beyond city-level initiatives. The country aims to deploy a 2,800-satellite constellation dedicated to space-based AI inference computing. Commercial operations are targeted for 2030, with the full network complete by 2035. More than 95 percent of the satellites will be dedicated to inference computing, processing AI workloads in orbit instead of on the ground.
The rationale is straightforward. Terrestrial data centers face mounting constraints in energy consumption, heat dissipation, and land availability as AI demand explodes. Space-based computing draws on solar power in orbit, covers wide areas, and can process raw data where it is collected — reducing the need for massive ground-to-space data transmissions.
“The construction of space-based computing infrastructure is significant,” said Xiang Ligang, director-general of the Zhongguancun Modern Information Consumer Application Industry Technology Alliance. “It will fundamentally alter the global supply pattern of computing power.”
National Support and Industrial Projections
The MIIT has formally pledged support for forward-looking research on space-based computing. The ministry is calling for the gradual establishment of a standards system covering hardware, software, networking and security, and is promoting R&D in space-borne radiation-resistant chips and inter-satellite laser communications.
According to preliminary estimates by the China Academy of Information and Communication Technology (CAICT), China’s space computing industry is expected to exceed 250 billion yuan ($36.6 billion) by 2030. Global industry forecasts are even more ambitious, with some analysts projecting the worldwide market will surpass $1 trillion by the same year.
Xie Lina, deputy director of CAICT’s Cloud Computing and Big Data Research Institute, noted that China is already at the forefront globally in space computing engineering and commercial deployment. But she cautioned that the industry remains in early stages and requires a pragmatic approach.
“To reduce the cost of launching satellites into orbit, the industry is accelerating efforts to develop reusable rocket technology and promote standardized, modular satellite design and mass production,” Xie said.
A Global Race
China is not alone in pursuing orbital compute. SpaceX has filed plans with the US Federal Communications Commission for a non-geostationary orbit system of up to 1 million satellites. Japan’s JAXA and the private firm iQPS have achieved in-orbit image processing of synthetic aperture radar data.
The institutional moves by Beijing, Tianjin and Chengdu represent a deliberate effort to coordinate the many pieces of orbital computing: launch, satellite manufacturing, chip design, AI software, and thermal management. Rather than leaving the sector to individual companies, China is building the regulatory and institutional frameworks to guide development at the national level.
The question now is whether orbital computing can deliver on its promise. The physics of heat dissipation in orbit, the cost of launching enough compute capacity, and the latency of space-ground links all remain unresolved challenges. But with multiple cities, companies, and government agencies now committed to finding answers, the race for space-based AI infrastructure has officially begun.