Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 13 min ago
PreprintCC BYarXiv (Cornell University)

Universal Machine-learning Molecular Dynamics at the Speed of Empirical Potentials

  • Tiancheng Li1,2
  • Jianming Xue3,2
  • Linfeng Zhang1,4
  • Duo Zhang
  • Han Wang3,5
5 affiliations
  1. 1AI for Science Institute , Beijing 100080 , P. R. China
  2. 2State Key Laboratory of Nuclear Physics and Technology , School of Physics , Peking University , Beijing 100871 , China
  3. 3HEDPS, CAPT , College of Engineering , Peking University , Beijing 100871 , P. R. China
  4. 4DP Technology , Beijing 100080 , P. R. China
  5. 5National Key Laboratory of Computational Physics , Institute of Applied Physics and Computational Mathematics , Fenghao East Road 2 , Beijing 100094 , P. R. China
Open at sourcePDFDOICite

Abstract

No interatomic potential has offered universality across chemistry, near-first-principles accuracy and the speed of empirical potentials at once. Here we introduce DPA4C, an equivariant potential whose architecture and compressed CUDA operators are co-designed under deployment constraints to pursue accuracy and efficiency together. Five variants spanning a 49-fold parameter range form the high-throughput end of the measured accuracy--throughput frontier. The largest variant approaches the accuracy of the MACE-Omat models at about two orders of magnitude higher measured throughput. The most compact reduces the energy, force and stress errors of the fastest existing universal MLIP by 61.4%, 48.1% and 34.3% at 1.92 times its saturated throughput. All five variants complete multimillion-atom simulations on a single GPU and run molecular dynamics for 2.048 billion atoms on 1,024 16-GB NVIDIA V100 GPUs at 83.3--91.2% weak-scaling efficiency. Compared with the MEAM empirical potential, DPA4C-Nano reaches 1.8 and 2.5 times the saturated throughput in single-GPU scans on the same V100 hardware for diamond carbon and FCC copper, respectively. DPA4C therefore brings quantum-trained universal accuracy into a regime of speed and system size previously associated with empirical potentials.

Related

  1. Correcting the Liquid Densities of Universal Machine-Learning Interatomic Potentials Using Local-Number-Density Repulsion
    Shunsuke Mieda
    ChemRxivPreprintLicence not statedGeneral chemistryChemRxiv
  2. Applicability Domain of Universal Machine Learning Interatomic Potentials at Solid-State Battery Interfaces
    Umang Agarwal, Mengke Li, Gopalakrishnan Sai Gautam +1
    ChemRxivPreprintLicence not statedGeneral chemistryChemRxiv
  3. Reference Conventions and Structure Types, Rather Than 4f Occupancy, Shape Lanthanide Errors in Universal Machine-Learning Interatomic Potentials
    Jungko Moni Chakma, Md. Mynul Hasan, Mohammad Asaduzzaman Chowdhury
    ChemRxivPreprintLicence not statedGeneral chemistryChemRxiv
  4. Fine-tuning universal machine learning potentials for transition state search in surface catalysis
    Raffaele Cheula, Mie Andersen, John R. Kitchin
    npj Computational MaterialsPeer-reviewedCC BY-NC-NDFull text7 citedMaterials & polymersOpenAlex
  5. Can Universal Machine-Learning Interatomic Potentials Capture Small Polarons? A Geometry-Based Benchmark in Transition-Metal Oxides
    Vladislav Korostelev, Pjotrs Žguns, Konstantin Klyukin
    arXivPreprintarXiv licenceFull textPhysical chemistryComputational physicsarXiv

Cite

BibTeXRIS
BibTeX
@misc{li2026universal,
  title = {{Universal Machine-learning Molecular Dynamics at the Speed of Empirical Potentials}},
  author = {Tiancheng Li and Jianming Xue and Linfeng Zhang and Duo Zhang and Han Wang},
  howpublished = {arXiv (Cornell University)},
  year = {2026},
  month = aug,
  doi = {10.48550/arxiv.2608.19041},
  url = {https://arxiv.org/abs/2608.19041},
  note = {CC BY},
}
RIS
TY  - UNPB
TI  - Universal Machine-learning Molecular Dynamics at the Speed of Empirical Potentials
AU  - Tiancheng Li
AU  - Jianming Xue
AU  - Linfeng Zhang
AU  - Duo Zhang
AU  - Han Wang
PY  - 2026
DA  - 2026/08/19
PB  - arXiv (Cornell University)
DO  - 10.48550/arxiv.2608.19041
UR  - https://arxiv.org/abs/2608.19041
AB  - No interatomic potential has offered universality across chemistry, near-first-principles accuracy and the speed of empirical potentials at once. Here we introduce DPA4C, an equivariant potential whose architecture and compressed CUDA operators are co-designed under deployment constraints to pursue accuracy and efficiency together. Five variants spanning a 49-fold parameter range form the high-throughput end of the measured accuracy--throughput frontier. The largest variant approaches the accuracy of the MACE-Omat models at about two orders of magnitude higher measured throughput. The most compact reduces the energy, force and stress errors of the fastest existing universal MLIP by 61.4%, 48.1% and 34.3% at 1.92 times its saturated throughput. All five variants complete multimillion-atom simulations on a single GPU and run molecular dynamics for 2.048 billion atoms on 1,024 16-GB NVIDIA V100 GPUs at 83.3--91.2% weak-scaling efficiency. Compared with the MEAM empirical potential, DPA4C-Nano reaches 1.8 and 2.5 times the saturated throughput in single-GPU scans on the same V100 hardware for diamond carbon and FCC copper, respectively. DPA4C therefore brings quantum-trained universal accuracy into a regime of speed and system size previously associated with empirical potentials.
SN  - 2331-8422
ER  -