AION-1: Omnimodal Foundation Model for Astronomical Sciences
Liam Holden Parker, Francois Lanusse, Jeff Shen, Ollie Liu, Tom Hehir, Leopoldo Sarra, Lucas Thibaut Meyer, Micah Bowles
Abstract
While foundation models have shown promise across a variety of fields, astronomy lacks a unified framework for joint modeling across its highly diverse data modalities. In this paper, we present AION-1, the first large-scale multimodal foundation family of models for astronomy. AION-1 enables arbitrary transformations between heterogeneous data types using a two-stage architecture: modality-specific tokenization followed by transformer-based masked modeling of cross-modal token sequences. Trained on over 200M astronomical objects, AION-1 demonstrates strong performance across regression, classification, generation, and object retrieval tasks. Beyond astronomy, AION-1 provides a scalable blueprint for multimodal scientific foundation models that can seamlessly integrate heterogeneous combinations of real-world observations. Our model release is entirely open source, including the dataset, training script, and weights.
BibTeX
@inproceedings{
parker2025aion,
title={{AION}-1: Omnimodal Foundation Model for Astronomical Sciences},
author={Liam Holden Parker and Francois Lanusse and Jeff Shen and Ollie Liu and Tom Hehir and Leopoldo Sarra and Lucas Thibaut Meyer and Micah Bowles and Sebastian Wagner-Carena and Helen Qu and Siavash Golkar and Alberto Bietti and Hatim Bourfoune and Pierre Cornette and Keiya Hirashima and Geraud Krawezik and Ruben Ohana and Nicholas Lourie and Michael McCabe and Rudy Morel and Payel Mukhopadhyay and Mariel Pettee and Kyunghyun Cho and Miles Cranmer and Shirley Ho},
booktitle={The Thirty-ninth Annual Conference on Neural Information Processing Systems},
year={2025},
url={https://openreview.net/forum?id=6gJ2ZykQ5W}
}