MLIPAudit: A benchmarking tool for Machine Learned Interatomic Potentials

Leon Wehrhan, Lucien Walewski, Marie Bluntzer, Heloise Chomet, Christoph Brunken, Jules Tilly, Silvia Acosta-Gutierrez

Track: Track 1: Original Research/Position/Education/Attention Track

Keywords: Artificial Intelligence, Digital Biology, Systems Engineering, Machine Learned Interatomic Potentials, Benchmark

TL;DR: We introduce MLIPAudit, a standardized benchmarking suite that enables consistent evaluation and comparison of machine-learned interatomic potentials across diverse molecular systems and tasks relevant for industrial applications..

Abstract: Machine-learned interatomic potentials (MLIPs) promise to significantly advance atomistic simulations by delivering quantum-level accuracy for large molecular systems at a fraction of the computational cost of traditional electronic structure methods. While model hubs and categorisation efforts have emerged in recent years, it remains difficult to consistently discover, compare, and apply these models across diverse scenarios. The field still lacks a standardised and comprehensive framework for evaluating MLIP performance. We introduce MLIPAudit, an open, curated and modular benchmarking suite designed to assess the accuracy of MLIP models across a variety of application tasks. MLIPAudit offers a diverse collection of benchmark systems, including small organic compounds, molecular liquids, proteins and flexible peptides, along with pre-computed results for a range of pre-trained and published models. MLIPAudit also provides tools for users to evaluate their models using the same standardised pipeline. A continuously updated leaderboard tracks performance across benchmarks, enabling direct comparison on downstream tasks. By offering a unified and transparent reference framework for model validation and comparison, MLIPAudit aims to foster reproducibility, transparency, and community-driven progress in the development of MLIPs for complex molecular systems. The library is available on GitHub and on PyPI 14 under the Apache license 2.0.

Submission Number: 347