Geospatial AI has had a transformative influence on Earth observation and remote sensing fields, with advancements in deep learning algorithms, software and hardware technologies, and the availability of high-resolution geospatial data. This has enabled a next generation of Earth observation and remote sensing, providing new means for researching Earth’s surface at various scales. Geospatial AI also emphasizes the cognition of the geographical environment and integrates multi-modal data, but comes with a high computational cost. To address this, data-driven machine learning within a simple framework is desired.