This dataset was launched in October 2020 when FAIR and Carnegie Mellon University jointly launched the Open Catalyst Project. It is one of the world's largest catalyst simulation datasets, the Open Catalyst 2020 (OC20) Dataset. The dataset contains millions of molecular simulation data generated by DFT calculations, which can help researchers train AI models for predicting energy changes and kinetics of catalytic reactions. The project is committed to helping scientists discover new materials that can promote the development of clean energy more quickly. The related paper results are "The Open Catalyst 2020 (OC20) Dataset and Community Challenges".