--- title: IBM and NASA release an open-source lunar foundation model description: IBM and NASA have released the NASA-IBM Lunar Foundation Model on Hugging Face along with the first unified, machine-learning-ready lunar dataset, combining... canonical: https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: IBM and NASA release an open-source lunar foundation model | daily.dev og:description: IBM and NASA have released the NASA-IBM Lunar Foundation Model on Hugging Face along with the first unified, machine-learning-ready lunar dataset, combining... og:url: https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl og:image: https://api.daily.dev/og/posts/rIuBtQqNL.png og:image:alt: IBM and NASA release an open-source lunar foundation model og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # IBM and NASA release an open-source lunar foundation model **[The Next Web](https://daily.dev/sources/tnw)** · 4 min read · 0 upvotes · 0 comments ## Summary IBM and NASA have released the NASA-IBM Lunar Foundation Model on Hugging Face along with the first unified, machine-learning-ready lunar dataset, combining over 30 spatially aligned layers from nine instruments across four missions (LRO, GRAIL, and Japan's SELENE/Kaguya). The model is aimed at tasks like identifying subsurface ice in permanently shadowed craters, mapping craters for landing site selection, and detecting volcanic features, reducing ice-detection error by 23% versus the general-purpose SwinV2-B model while needing less compute. Part of IBM's Prithvi family (which also covers Earth, weather, and heliophysics), the release notably lacks disclosed model size, parameter count, license, or architecture details. It also arrives as Hugging Face itself is being acquired by Nvidia for $12.93bn. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the NASA-IBM Lunar Foundation Model and where can I access it? It is an open-source AI model trained on decades of lunar observation data, released by IBM and NASA on Hugging Face alongside the first unified, machine-learning-ready lunar dataset. The dataset combines more than 30 spatially aligned layers from nine instruments across four missions, including NASA's Lunar Reconnaissance Orbiter, GRAIL, and Japan's SELENE/Kaguya. It is part of IBM's Prithvi model family. _Researchers evaluating geospatial foundation models can follow model and dataset releases like this on daily.dev._ ### How accurate is the NASA-IBM Lunar Foundation Model at detecting lunar ice compared to other vision models? It reduces error in identifying high-ice-potential areas by 23% compared with SwinV2-B, a general-purpose vision model trained on ImageNet. Gains are smaller elsewhere: only 3% improvement for volcanic feature detection, and roughly equal performance to SwinV2-B for metre-scale crater detection, though with lower compute requirements. At around 100-metre resolution it outperforms SwinV2-B by nearly 19% while using half the training data. _Teams comparing geospatial models on accuracy versus compute cost can track benchmarks like these on daily.dev._ ### Why hasn't IBM and NASA disclosed the model size or license for the Lunar Foundation Model? No model size, parameter count, license, or architecture description has been provided at release, despite these being among the first details a research team needs before deciding whether to adapt the model. The release also coincides with Nvidia's agreement to acquire Hugging Face for $12.93bn, meaning the platform hosting the model is set to gain a commercial owner. _Developers deciding whether to build on an open model can watch for licensing details via daily.dev._ ## Similar posts on daily.dev - [Astrolab’s upcoming rover will run the first LLMs on the moon](https://daily.dev/posts/astrolab-s-upcoming-rover-will-run-the-first-llms-on-the-moon-rqdtdovty) · MIT Technology Review · 0 upvotes · 0 comments - [NVIDIA and Hugging Face Bring New Models and Frameworks to LeRobot for the Open Robotics Community](https://daily.dev/posts/nvidia-and-hugging-face-bring-new-models-and-frameworks-to-lerobot-for-the-open-robotics-community-dbghgulvq) · NVIDIA · 0 upvotes · 0 comments - [Cambridge's planet-scale AI model runs on AMD, not Nvidia](https://daily.dev/posts/cambridge-s-planet-scale-ai-model-runs-on-amd-not-nvidia-opszhyitu) · The Next Web · 0 upvotes · 0 comments - [State of Open Source on Hugging Face: Spring 2026](https://daily.dev/posts/state-of-open-source-on-hugging-face-spring-2026-fuzipfbpa) · Hugging Face · 2 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#open-source](https://daily.dev/tags/open-source), [#ibm](https://daily.dev/tags/ibm), [#nasa](https://daily.dev/tags/nasa) [View this post on daily.dev](https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"IBM and NASA release an open-source lunar foundation model","url":"https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl"},"datePublished":"2026-09-10T12:17:06.985Z","dateModified":"2026-09-14T06:58:39.234Z","description":"IBM and NASA have released the NASA-IBM Lunar Foundation Model on Hugging Face along with the first unified, machine-learning-ready lunar dataset, combining...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c2f6b1cffba1d0c524fa7c6580bd282a?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/c2f6b1cffba1d0c524fa7c6580bd282a?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"The Next Web","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"The Next Web","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/tnw","url":"https://daily.dev/sources/tnw"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,open-source,ibm,nasa","timeRequired":"PT4M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"The Next Web","item":"https://daily.dev/sources/tnw"},{"@type":"ListItem","position":3,"name":"IBM and NASA release an open-source lunar foundation model"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ibm-and-nasa-release-an-open-source-lunar-foundation-model-riubtqqnl#faq","mainEntity":[{"@type":"Question","name":"What is the NASA-IBM Lunar Foundation Model and where can I access it?","acceptedAnswer":{"@type":"Answer","text":"It is an open-source AI model trained on decades of lunar observation data, released by IBM and NASA on Hugging Face alongside the first unified, machine-learning-ready lunar dataset. The dataset combines more than 30 spatially aligned layers from nine instruments across four missions, including NASA's Lunar Reconnaissance Orbiter, GRAIL, and Japan's SELENE/Kaguya. It is part of IBM's Prithvi model family. Researchers evaluating geospatial foundation models can follow model and dataset releases like this on daily.dev."}},{"@type":"Question","name":"How accurate is the NASA-IBM Lunar Foundation Model at detecting lunar ice compared to other vision models?","acceptedAnswer":{"@type":"Answer","text":"It reduces error in identifying high-ice-potential areas by 23% compared with SwinV2-B, a general-purpose vision model trained on ImageNet. Gains are smaller elsewhere: only 3% improvement for volcanic feature detection, and roughly equal performance to SwinV2-B for metre-scale crater detection, though with lower compute requirements. At around 100-metre resolution it outperforms SwinV2-B by nearly 19% while using half the training data. Teams comparing geospatial models on accuracy versus compute cost can track benchmarks like these on daily.dev."}},{"@type":"Question","name":"Why hasn't IBM and NASA disclosed the model size or license for the Lunar Foundation Model?","acceptedAnswer":{"@type":"Answer","text":"No model size, parameter count, license, or architecture description has been provided at release, despite these being among the first details a research team needs before deciding whether to adapt the model. The release also coincides with Nvidia's agreement to acquire Hugging Face for $12.93bn, meaning the platform hosting the model is set to gain a commercial owner. Developers deciding whether to build on an open model can watch for licensing details via daily.dev."}}]} ```