Uni-Mol is the first universal large-scale 3D Molecular Representation Learning framework developed by DP Technology, with two models trained on billions of molecular 3D conformations and millions of protein pocket data. It delivers strong performance in molecular property prediction, especially 3D-related tasks, and can also predict material properties such as MOF gas adsorption and OLED molecular optics. The project is MIT-licensed and provides tools, tutorials, and utilities (Uni-Mol tools, data interfaces, training workflows) to enable property prediction, representation, and downstream tasks.
AI named Uni-Mol in September 2026.
Brand page for unimol-readthedocs-io