Are Molecular Foundation Models Ready for Free Energy Simulations?
- 1Bavarian Center for Battery Technology (BayBatt
- 2Fritz Haber Institute of the Max Planck Society
Abstract
Foundation models for atomistic simulations offer broadly applicable potentialenergy surfaces that can either be applied in a zero-shot manner (without domainspecific retraining) or fine-tuned to the task at hand. Their accuracy is typically judged by energy and force errors on near-equilibrium data, yet it remains largely unclear whether these metrics translate into reliable free-energy landscapes. Here, enhanced sampling can be a particularly demanding test: simulations are driven far from equilibrium into sparsely trained regions and held there over long trajectories, so that small systematic errors in forces or relative energies can accumulate and distort predicted populations, transition pathways, and barriers. We assess several molecular foundation models trained on different dataset sizes and architectures (including models with explicit long-range electrostatics) in prototypical metadynamics and umbrella sampling simulations. Our results show that recent foundation models can reproduce simple gas-phase free-energy profiles with near-reference accuracy, while more complex multidimensional and solvated free-energy landscapes remain challenging. However, assessing their performance in these systems is further complicated by limitations in current sampling protocols and the lack of well-defined reference data. These findings highlight the need for more rigorous sampling protocols and reliable benchmarks to systematically assess the accuracy of future molecular foundation models in free-energy calculations.
Related
- Teleported free energy simulations
- Foundation Neural-Network Quantum States for Molecular Potential Energy Surfaces in Second Quantization
- Model-free Reconstruction of Molecular Energy Levels by Broadband Kilohertz-accurate Cavity-enhanced Spectroscopy
- Unsupervised Thermodynamics of Molecular Diffusion Models: Action-Operator Semantics and Auditable Free-Energy Readout
- Bicarbonate/carbonic acid equilibria in aqueous solutions: Conformational Dynamics, Ion-Pair Speciation, and Free Energy Landscapes from Ab Initio Molecular Dynamics and Metadynamics
BibTeX
@misc{gonnheimer2026molecular,
title = {{Are Molecular Foundation Models Ready for Free Energy Simulations?}},
author = {Nils Gönnheimer and Karsten Reuter and Johannes T. Margraf},
howpublished = {ChemRxiv},
year = {2026},
month = oct,
doi = {10.26434/chemrxiv.15010593/v1},
url = {https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010593/v1},
note = {CC BY},
}
RIS
TY - UNPB
TI - Are Molecular Foundation Models Ready for Free Energy Simulations?
AU - Nils Gönnheimer
AU - Karsten Reuter
AU - Johannes T. Margraf
PY - 2026
DA - 2026/10/16
PB - ChemRxiv
DO - 10.26434/chemrxiv.15010593/v1
UR - https://chemrxiv.org/doi/full/10.26434/chemrxiv.15010593/v1
AB - Foundation models for atomistic simulations offer broadly applicable potentialenergy surfaces that can either be applied in a zero-shot manner (without domainspecific retraining) or fine-tuned to the task at hand. Their accuracy is typically judged by energy and force errors on near-equilibrium data, yet it remains largely unclear whether these metrics translate into reliable free-energy landscapes. Here, enhanced sampling can be a particularly demanding test: simulations are driven far from equilibrium into sparsely trained regions and held there over long trajectories, so that small systematic errors in forces or relative energies can accumulate and distort predicted populations, transition pathways, and barriers. We assess several molecular foundation models trained on different dataset sizes and architectures (including models with explicit long-range electrostatics) in prototypical metadynamics and umbrella sampling simulations. Our results show that recent foundation models can reproduce simple gas-phase free-energy profiles with near-reference accuracy, while more complex multidimensional and solvated free-energy landscapes remain challenging. However, assessing their performance in these systems is further complicated by limitations in current sampling protocols and the lack of well-defined reference data. These findings highlight the need for more rigorous sampling protocols and reliable benchmarks to systematically assess the accuracy of future molecular foundation models in free-energy calculations.
ER -