IBM and NASA have released the NASA-IBM Lunar Foundation Model, an open-source artificial intelligence system designed to help scientists study the Moon’s surface more efficiently. Announced on September 10, 2026, the model is available as a free download on Hugging Face, with its full codebase published on GitHub.
The model was trained on 17 years of lunar data collected by nine instruments across four space missions, including NASA’s Lunar Reconnaissance Orbiter, NASA’s GRAIL mission, and Japan’s SELENE/Kaguya mission. It can detect potential water ice deposits in permanently shadowed regions, map volcanic features known as Irregular Mare Patches, and identify craters. According to IBM and NASA, it outperformed the widely used SwinV2-B vision model by up to 23% in locating ice deposits, by nearly 19% in crater detection at context-scale resolution, and by 3% in capturing volcanic formations, while using only half the training data for some tasks.
Alongside the model, the organizations released what they describe as the first open-source unified lunar dataset of its kind. It brings together more than 30 spatially aligned layers from the four missions into a machine-learning-ready format, including tens of thousands of images and maps of the Moon’s surface and subsurface. IBM Research Europe director Juan Bernabe-Moreno said the roughly two million co-registered data points may prove to be the most lasting contribution, adding, ‘The data is what really creates the industry of AI models.’ NASA chief science data officer Kevin Murphy said the release is intended to make decades of NASA data easier for scientists to explore.
The model was also tested on August 5, 2026, when IBM fed it an image showing the impact site left after a SpaceX Falcon 9 rocket struck the Moon. Despite the new impact appearing almost directly over an existing crater, the model correctly identified it as a newly formed crater. The lunar model is the latest addition to IBM’s Prithvi family of open foundation models, which already covers geospatial, weather, and heliophysics applications. IBM stock was trading down about 1.38% at the time of the announcement.
