Abstract: This study combines artificial intelligence and deep learning technologies to construct a drug solubility prediction platform within the Uni-Mol, offering a replicable cloud-based teaching solution. Leveraging cloud computing and interactive experiments, the platform enables instructors to rapidly establish customized teaching environments while guiding students in conducting solubility predictions through data analysis and model inference. The modular experimental design encompasses multi-level tasks, ranging from knowledge learning to advanced skill development, including SMILES parsing, molecular structure visualization, model training, and evaluation. This approach not only enhances students’ interdisciplinary competencies in computational chemistry and artificial intelligence (AI) but also facilitates the practical application of AI technology in chemistry education.

Key words: Solubility Prediction, Uni-Mol, Cloud-based Teaching, Artificial Intelligence