A Long March-2D carrier rocket carrying a space computing satellite constellation blasts off from the Jiuquan Satellite Launch Center in northwest China on May 14, 2025. Photo: Xinhua

China's space start-up ADA Space on Thursday unveiled its plan for a new computing satellite constellation of 12 satellites, with computing power expected to be four times greater than the inaugural batch launched in May. The project accelerates the construction of its Star Compute program, which will ultimately comprise 2,800 supercomputer satellite systems. 

At a ceremony held in Wuxi, East China's Jiangsu Province on Thursday, ADA Space revealed that the new constellation, codenamed Liangxi, will consist of 12 intelligent computing satellites and have a combined computing power of 20 peta operations per second (POPS).

The first tranche of computing satellites of the Star Compute program was successfully launched on May 14 via a Long March-2D carrier rocket from the Jiuquan Satellite Launch Center in Northwest China. It also marked the launch of the world's first space computing constellation.

The initial satellites provide a combined computing power of 5 POPS and 30 terabytes of storage capacity. They are equipped with onboard intelligent computing and inter-satellite communication systems, and are capable of in-orbit computing, Xinhua News Agency reported in May. 

Currently, the development and deployment of the second and third batches of computing satellites of the program are also underway, ADA Space said Thursday. 

The Star Compute program officially delivered services to its first client, the Aerospace Information Research Institute (AIRI) of the Chinese Academy of Sciences (CAS), on Thursday, marking a key step toward real-world application of its core space computing capabilities. 

ADA Space told the Global Times on Thursday that AIRI's optical satellite target detection algorithm model successfully deployed and performed real-time inference tasks on the on-orbit AI computing payload, verifying the compatibility of the on-orbit AI computing payload with the remote sensing intelligent detection model and providing important reference for subsequent on-orbit AI applications.

While the US and Europe have tested edge computing in space, China's collaborative effort appears to be the first to deploy a purpose-built AI constellation at scale, Andrew Jones, a China space observer wrote in his article published on spacenews.com in May.

According to Zhao Hongjie, executive vice president of ADA Space, these computing satellites are not only significant for advancing space technologies - such as supporting spacecraft in low-Earth orbit, lunar, and Mars missions - but also open up new possibilities for testing large-scale computing architectures and serving future applications on Earth and in low-altitude airspace. The satellites can process collected data onboard instead of transmitting it to Earth, greatly increasing efficiency, Zhao told media in May. 

At the same time, the low-temperature environment and abundant solar energy in space provide natural advantages for achieving low-energy, sustainable space computing, Wang Yanan, chief editor of Aerospace Knowledge magazine, told the Global Times. 

However, space computing also faces challenges. The harsh space environment, including near-vacuum conditions, high radiation, microgravity, and extreme temperature fluctuations, requires specialized materials and radiation-hardened components to ensure reliability and longevity, space observers noted.