# lunar foundation model

Open-source AI foundation model for lunar-science data analysis, with a public fine-tuning and inference software release.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## NASA-IBM Lunar Foundation Model Goes Open Source With a 2M-Tile Dataset and 22% Lower Ice-Mapping Error

DevFeed: [NASA-IBM Lunar Foundation Model Goes Open Source With a 2M-Tile Dataset and 22% Lower Ice-Mapping Error](<https://devfeed.tech/articles/nasa-ibm-lunar-foundation-model-goes-open-source-with-a-2m-tile-dataset-and-22-lower-ice-mapping-error-17437.md>)

Original publisher: [Read original article](<https://www.storagereview.com/news/nasa-ibm-lunar-foundation-model-goes-open-source-with-a-2m-tile-dataset-and-22-lower-ice-mapping-error>)

Author: Harold Fritts

Published: 2026-09-14T16:43:16Z

Content type: news

Language: en

Sources: [StorageReview.com](<https://devfeed.tech/sources/storagereview-com.md>)

Topics: [lunar foundation model](<https://devfeed.tech/topics/lunar-foundation-model.md>), [foundation-models](<https://devfeed.tech/topics/foundation-models.md>), [ibm](<https://devfeed.tech/topics/ibm.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [hugging face](<https://devfeed.tech/topics/hugging-face.md>), [dataset](<https://devfeed.tech/topics/dataset.md>), [multimodal](<https://devfeed.tech/topics/multimodal.md>), [Benchmark](<https://devfeed.tech/topics/benchmark.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [data](<https://devfeed.tech/tags/data.md>), [enterprise](<https://devfeed.tech/tags/enterprise.md>), [fine-tuning](<https://devfeed.tech/tags/fine-tuning.md>), [foundation-models](<https://devfeed.tech/tags/foundation-models.md>), [hugging-face](<https://devfeed.tech/tags/hugging-face.md>), [ibm](<https://devfeed.tech/tags/ibm.md>), [lunar-foundation-model](<https://devfeed.tech/tags/lunar-foundation-model.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [nasa](<https://devfeed.tech/tags/nasa.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [space](<https://devfeed.tech/tags/space.md>)

### AI overview

IBM and NASA have released the NASA-IBM Lunar Foundation Model as open source on Hugging Face, along with its weights, technical report, and training dataset. Built on TerraMind, the model uses multimodal lunar observations for tasks including ice-deposit mapping, volcanic-feature detection, and crater detection. Reported benchmarks show up to 22% lower ice-mapping error than SwinV2-B, while the accompanying dataset contains roughly 2 million image tiles from nine instruments across four lunar missions.

### Source excerpt

IBM and NASA have released the NASA-IBM Lunar Foundation Model as open source, one of the first publicly available foundation models built for scientific study of the Moon. The weights, a technical report, and the machine-learning-ready dataset it was trained on are up on Hugging Face under the Prithvi family, which already covers Earth observation, The post NASA-IBM Lunar Foundation Model Goes Open Source With a 2M-Tile Dataset and 22% Lower Ice-Mapping Error appeared first on StorageReview.com.

## Introducing IBM and NASA's new foundation model for the Moon

DevFeed: [Introducing IBM and NASA's new foundation model for the Moon](<https://devfeed.tech/articles/introducing-ibm-and-nasa-s-new-foundation-model-for-the-moon-17342.md>)

Original publisher: [Read original article](<https://research.ibm.com/blog/nasa-ibm-lunar-foundation-model>)

Author: Kim Martineau

Published: 2026-09-10T12:30:00Z

Content type: article

Language: en

Sources: [IBM Research](<https://devfeed.tech/sources/ibm-research.md>)

Topics: [lunar foundation model](<https://devfeed.tech/topics/lunar-foundation-model.md>), [ibm](<https://devfeed.tech/topics/ibm.md>), [foundation-models](<https://devfeed.tech/topics/foundation-models.md>), [data](<https://devfeed.tech/topics/data.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI Architecture](<https://devfeed.tech/topics/ai-architecture.md>)

Tags: [accelerated-discovery](<https://devfeed.tech/tags/accelerated-discovery.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-architecture](<https://devfeed.tech/tags/ai-architecture.md>), [algorithms](<https://devfeed.tech/tags/algorithms.md>), [ibm](<https://devfeed.tech/tags/ibm.md>), [lunar-foundation-model](<https://devfeed.tech/tags/lunar-foundation-model.md>), [mapping](<https://devfeed.tech/tags/mapping.md>), [model](<https://devfeed.tech/tags/model.md>), [nasa](<https://devfeed.tech/tags/nasa.md>), [release](<https://devfeed.tech/tags/release.md>), [science](<https://devfeed.tech/tags/science.md>), [space](<https://devfeed.tech/tags/space.md>), [us](<https://devfeed.tech/tags/us.md>)

### AI overview

IBM and NASA are open-sourcing the NASA-IBM Lunar Foundation Model, a multimodal AI model that integrates lunar observations from US and Japanese missions across viewing angles, spatial scales, and measurement types. The model is intended to support lunar mapping, volcanic-history research, and searches for polar ice.

### Source excerpt

The multi-modal model could help astronauts navigate craters, investigate ancient lava, and search for ice, as the US plans for a long-term lunar presence.