LunarFM
A multimodal foundation model for lunar science. Trained on orbital imagery, elevation, gravity, and spectral data to enable intelligent exploration of the Moon.
Paper
Read the preprint to learn more about how we put together this first version of LunarFM.
Data
Access the LunarChips dataset — curated input data for training your own lunar foundation model.
Embeddings
Use pre-computed LunarEmbeddings directly for downstream tasks without retraining from scratch.
Model
Download the pre-trained LunarFM model weights to run inference or fine-tune on your own lunar datasets.
Codebase
Explore the full source code on GitHub to understand the architecture and reproduce our results.
Tutorials
Get started quickly with our tutorial notebooks on Colab covering inference, fine-tuning, and data loading.
🚀 Links to the data, model, embeddings, code & tutorials are coming soon — stay tuned!