Machine Learning Potential

A machine learning potential (MLP) learns the relationship between atomic structures and their energies and forces from quantum mechanical data, such as DFT calculations. Once trained, it can evaluate new structures much faster than DFT, often with near-DFT accuracy. An online MLP demo is available for exploring the FAIR Chem Universal Materials Accelerator (UMA) model in your browser.

Online Tool site: FAIR Chem UMA Demo

Key Features and Advantages of UMA

UMA models predict motion and behavior at the atomic scale, ultimately reducing the development cycle in molecular and materials discovery and unlocking new possibilities for innovation and impact.
UMA models are based on Density Functional Theory (DFT) training datasets. DFT simulations are a commonly used quantum chemistry method to simulate and understand behavior at the atomic scale.

User Guide & Workflow

Step 1.

Go to Explore UMA's capabilities

Navigate to the "2. Explore UMA's capabilities" tab at the top of the demo interface.

Step 2.

Select a Pre-loaded Sample Structure

Choose any preset material or system from the list

(e.g., Transition metal complexes, Organic molecular crystal, Inoganic crystal).

Step 3.

View UMA Simulation Results

Inspect the real-time 3D atomic structure under the "Visualization" panel

Check the generated calculation Log and Script.

Page updated

Google Sites

Report abuse