Zatom-1: Towards a Multimodal Foundation Model for 3D Molecules and Materials

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Morehead, Alex, Cretu, Miruna, Panescu, Antonia, Anand, Rishabh, Weiler, Maurice, Perez, Tynan, Blau, Samuel, Farrell, Steven, Bhimji, Wahid, Jain, Anubhav, Sahasrabuddhe, Hrushikesh, Lio, Pietro, Jaakkola, Tommi, Gomez-Bombarelli, Rafael, Ying, Rex, Erichson, N. Benjamin, Mahoney, Michael W.
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866910216958771200
author Morehead, Alex
Cretu, Miruna
Panescu, Antonia
Anand, Rishabh
Weiler, Maurice
Perez, Tynan
Blau, Samuel
Farrell, Steven
Bhimji, Wahid
Jain, Anubhav
Sahasrabuddhe, Hrushikesh
Lio, Pietro
Jaakkola, Tommi
Gomez-Bombarelli, Rafael
Ying, Rex
Erichson, N. Benjamin
Mahoney, Michael W.
author_facet Morehead, Alex
Cretu, Miruna
Panescu, Antonia
Anand, Rishabh
Weiler, Maurice
Perez, Tynan
Blau, Samuel
Farrell, Steven
Bhimji, Wahid
Jain, Anubhav
Sahasrabuddhe, Hrushikesh
Lio, Pietro
Jaakkola, Tommi
Gomez-Bombarelli, Rafael
Ying, Rex
Erichson, N. Benjamin
Mahoney, Michael W.
contents General-purpose 3D modeling in chemistry encompasses molecules and materials, requiring both generative and predictive capabilities. However, most existing AI approaches are optimized for a single domain (molecules or materials) and a single task (generation or prediction), which limits representation sharing and transfer. We introduce Zatom-1, a cross-domain, general-purpose model architecture that unifies generative and predictive learning of 3D molecules and materials. Zatom-1 is a deliberately simplified Transformer trained with a multimodal flow matching objective that jointly models discrete atom types and continuous 3D geometries. This approach supports scalable pretraining with predictable gains as model capacity increases, while enabling fast and stable sampling. We use cross-domain generative pretraining as a universal initialization for downstream multi-task prediction of properties, energies, and forces. Empirically, Zatom-1 outperforms or competes with specialized baselines on both multi-task generative and predictive benchmarks in data-controlled settings, while improving generative inference speed by more than an order of magnitude. Our experiments demonstrate positive predictive transfer between data domains from joint generative pretraining: modeling materials during generative pretraining improves molecular property prediction accuracy. Open-source code and model weights are freely available at https://github.com/Zatom-AI/zatom.
format Preprint
id arxiv_https___arxiv_org_abs_2602_22251
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Zatom-1: Towards a Multimodal Foundation Model for 3D Molecules and Materials
Morehead, Alex
Cretu, Miruna
Panescu, Antonia
Anand, Rishabh
Weiler, Maurice
Perez, Tynan
Blau, Samuel
Farrell, Steven
Bhimji, Wahid
Jain, Anubhav
Sahasrabuddhe, Hrushikesh
Lio, Pietro
Jaakkola, Tommi
Gomez-Bombarelli, Rafael
Ying, Rex
Erichson, N. Benjamin
Mahoney, Michael W.
Machine Learning
Materials Science
Artificial Intelligence
General-purpose 3D modeling in chemistry encompasses molecules and materials, requiring both generative and predictive capabilities. However, most existing AI approaches are optimized for a single domain (molecules or materials) and a single task (generation or prediction), which limits representation sharing and transfer. We introduce Zatom-1, a cross-domain, general-purpose model architecture that unifies generative and predictive learning of 3D molecules and materials. Zatom-1 is a deliberately simplified Transformer trained with a multimodal flow matching objective that jointly models discrete atom types and continuous 3D geometries. This approach supports scalable pretraining with predictable gains as model capacity increases, while enabling fast and stable sampling. We use cross-domain generative pretraining as a universal initialization for downstream multi-task prediction of properties, energies, and forces. Empirically, Zatom-1 outperforms or competes with specialized baselines on both multi-task generative and predictive benchmarks in data-controlled settings, while improving generative inference speed by more than an order of magnitude. Our experiments demonstrate positive predictive transfer between data domains from joint generative pretraining: modeling materials during generative pretraining improves molecular property prediction accuracy. Open-source code and model weights are freely available at https://github.com/Zatom-AI/zatom.
title Zatom-1: Towards a Multimodal Foundation Model for 3D Molecules and Materials
topic Machine Learning
Materials Science
Artificial Intelligence
url https://arxiv.org/abs/2602.22251