What is MBX?
MBX is a C++ library that provides energies and forces for data-driven many-body potential energy functions (PEFs), including:
- MB-pol for water
- MB-nrg for halide and alkali-metal ions
- CO₂, CH₄, NH₃ and mixed gas-phase / condensed-phase systems
- Covalently bonded molecules (e.g., alkanes, N-methyl acetamide)
MBX interfaces with molecular-dynamics engines such as:
- LAMMPS (recommended for classical MD due to performance)
- i-PI (recommended for path-integral MD)
MBX acts as a client that returns energies and forces, while the MD engine controls the trajectory. The library supports:
- OpenMP and MPI parallelism
- Classical, quantum, and hybrid MD
- Periodic & nonperiodic simulations
- JSON-based configuration
- Many-body expansions via the MB-nrg formalism
Tutorials
The MBX tutorials include:
- Compilation and installation
- Testing and benchmark workflows
- JSON configuration
- Stand-alone executables
- LAMMPS and i-PI interfaces
- Using MB-nrg / MB-pol with external force fields
Online Tutorials
Updates
Recent versions:
- MBX v1.3.0 — 2025-07-09
- MBX v1.2.0 — 2024-08-20
- MBX v1.1.0 — 2024-04-22
All performance improvements and new MB-nrg PEFs are listed on the Updates page.
Team & Contributors
MBX is developed and maintained by the Paesani Research Group, UC San Diego. Current contributors include Xuanyu Zhu, Richa Rashmi, Henry Agnew, Ethan Bull-Vulpe, Xinmao Lyu, Ruihan Zhou, Yuehua Xie, and Shishir Iyer (see About Us page).
FAQ
MBX is continuously updated with new PEFs and performance improvements. For questions about installation, LAMMPS / i-PI interfaces, runtime errors, or slow-performance diagnostics, consult the FAQ or post on the MBX Google Group.
Download
The latest version of MBX can be downloaded from GitHub:
MBX GitHub Releases
Key Publications
- MBX v1.2: Accelerating data-driven many-body MD simulations. S. Gupta, E. F. Bull-Vulpe, H. Agnew, S. Iyer, X. Zhu, R. Zhou, C. Knight, F. Paesani, J. Chem. Theory Comput. 21, 1838 (2025). doi:10.1021/acs.jctc.4c01333
- MBX: A many-body energy and force calculator for data-driven many-body simulations. M. Riera, C. Knight, E. F. Bull-Vulpe, X. Zhu, H. Agnew, D. G. A. Smith, A. C. Simmonett, F. Paesani, J. Chem. Phys. 159, 054802 (2023). doi:10.1063/5.0156036
