# Researchers unlock machine learning potential with 120 million atomic configurations for materials discovery

**URL:** <https://community.materialssquare.com/t/researchers-unlock-machine-learning-potential-with-120-million-atomic-configurations-for-materials-discovery/1203>\
**Category:** News\
**Created:** [September 5, 2025, 5:29am UTC](https://community.materialssquare.com/t/researchers-unlock-machine-learning-potential-with-120-million-atomic-configurations-for-materials-discovery/1203 "2025-09-05T05:29:53Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![Ahn\_Yurim](https://avatars.discourse-cdn.com/v4/letter/a/c5a1d2/32.png) [@Ahn\_Yurim](https://community.materialssquare.com/u/Ahn_Yurim)\
**Post date:** [September 5, 2025, 5:29am UTC](https://community.materialssquare.com/t/researchers-unlock-machine-learning-potential-with-120-million-atomic-configurations-for-materials-discovery/1203/1 "2025-09-05T05:29:53Z")

</div>

The LeMat-Traj dataset, with 120 million atomic configurations, accelerates materials discovery via machine learning. Standardized by DFT calculations, it reduces energy/force prediction errors by 36%. It improves Matbench Discovery stability by 10%, revolutionizing materials science research.

For more details, please continue reading the full article under the following link:

> **[Researchers Unlock Machine Learning Potential With 120 Million Atomic...](https://quantumzeitgeist.com/researchers-unlock-machine-learning-potential-with-120-million-atomic-configurations-for-materials-discovery/)**
>
> Researchers have created a comprehensive, standardised dataset of over 120 million atomic configurations, called LeMat-Traj, which dramatically simplifies the development of more accurate machine learning models for predicting material behaviour

The Materials Square platform in fact provides extensive web-based functionalities for computational chemistry/materials research and training/education purposes, as detailed also in the following PDF brochure:  
[MaterialsSquare-brochure(2024).pdf](https://community.materialssquare.com/uploads/short-url/gpMFGwPXItuGUXGHG2sA2hBJCuY.pdf) (5.0 MB)

#materials #materialsscience #materialsengineering #computationalchemistry #modelling #chemistry #researchanddevelopment #research #MaterialsSquare #ComputationalChemistry #Tutorial #DFT #simulationsoftware #simulation