On April 28, 2026, the “Panshi Linkong” (磐石·临空) domain-specific AI large model was officially unveiled at the Chinese Academy of Sciences’ “Panshi 100” model system launch event. Developed jointly by the Aerospace Information Research Institute and the Institute of Automation of CAS, it is China’s first AI model dedicated to near-space domain knowledge reconstruction and intelligent research.

Near space, typically defined as the region between 20 and 100 kilometers above Earth’s surface, is a critical zone bridging aviation and aerospace. It holds strategic importance for communications relay, weather monitoring, and national defense. However, research in this field has long lacked specialized intelligent analysis tools due to the complexity of the data and high technical barriers.

The release of “Panshi Linkong” marks a significant milestone in China’s efforts to digitize and advance near-space research. The model integrates domain knowledge from atmospheric physics,飞行器 design, and space environment sciences, enabling intelligent analysis and decision support for complex near-space systems.

This release represents an important expansion of CAS’s “Panshi” scientific foundation model system. The original “Panshi Scientific Foundation Model” was launched at the 2025 World AI Conference, covering six core disciplines: mathematics, physics, chemistry, astronomy, geography, and biology. “Panshi Linkong” extends this ecosystem with specialized capabilities for the near-space domain.

Industry experts believe the model will accelerate scientific innovation in near-space research and provide critical intelligent support for China’s aerospace strategy.

Source: China News Service, Chinese Academy of Sciences