IBM (NYSE: IBM) and NASA have released the NASA-IBM Lunar Foundation Model, an open-source artificial intelligence model designed to help scientists analyze lunar surface data, according to a press release issued Sept. 10, 2026.

The model was trained on a dataset compiled by IBM and NASA researchers, aggregating more than 30 spatially aligned data layers from nine instruments across four missions, including NASA's Lunar Reconnaissance Orbiter, NASA's GRAIL mission, and the Japan Aerospace Exploration Agency's SELENE/Kaguya spacecraft.

According to a technical paper authored by IBM and NASA, the model outperforms the SwinV2-B ImageNet model by up to 23% across several tasks, including identifying potential lunar ice deposits, mapping volcanic formations known as Irregular Mare Patches, and detecting craters. Specifically, the model reduced error in identifying areas with high potential for lunar ice by up to 22%, and outperformed the same benchmark model in crater detection at context-scale resolution by nearly 19% using half the training data.

Kevin Murphy, chief science data officer and acting chief data and AI officer at NASA Headquarters, said in the release: "The NASA-IBM Lunar Foundation Model shows what's possible when we bring AI to NASA's petabytes of scientific data."

Juan Bernabe-Moreno, Director of IBM Research Europe, UK and Ireland, stated: "The NASA-IBM Lunar Foundation Model gives scientists a foundation to explore the Moon at scale, connecting observations across instruments, revealing patterns that are difficult to see in isolation."

The model joins IBM's Prithvi family of open foundation models, which also covers geospatial, weather, and heliophysics applications. The accompanying lunar dataset is described by the companies as the first open-source unified machine learning-ready lunar dataset of its kind.