The paper introduces SolPredictor, a computational model designed to predict molecular solubility using a residual gated graph neural network (RGNN). The model aims to enhance drug discovery by providing rapid and accurate assessments of compound properties before costly laboratory experiments. SolPredictor is based on RGNNs, which are designed to capture long-range dependencies in graph-structured data. The model uses a simplified molecular-input line-entry system (SMILES) representation and is evaluated using ten-fold cross-validation, achieving an R² of 0.79 ± 0.02 and an RMSE of 1.03 ± 0.04. The model was tested on five independent datasets, demonstrating its effectiveness in predicting solubility. Error analysis, hyperparameter optimization, and model explainability were conducted to identify the most valuable molecular features for prediction. The paper also discusses the development of a web server for SolPredictor, which can be used by researchers and pharmaceutical industry experts to predict solubility based on SMILES input. The study highlights the importance of solubility in drug discovery and the potential of computational methods to improve drug development efficiency.The paper introduces SolPredictor, a computational model designed to predict molecular solubility using a residual gated graph neural network (RGNN). The model aims to enhance drug discovery by providing rapid and accurate assessments of compound properties before costly laboratory experiments. SolPredictor is based on RGNNs, which are designed to capture long-range dependencies in graph-structured data. The model uses a simplified molecular-input line-entry system (SMILES) representation and is evaluated using ten-fold cross-validation, achieving an R² of 0.79 ± 0.02 and an RMSE of 1.03 ± 0.04. The model was tested on five independent datasets, demonstrating its effectiveness in predicting solubility. Error analysis, hyperparameter optimization, and model explainability were conducted to identify the most valuable molecular features for prediction. The paper also discusses the development of a web server for SolPredictor, which can be used by researchers and pharmaceutical industry experts to predict solubility based on SMILES input. The study highlights the importance of solubility in drug discovery and the potential of computational methods to improve drug development efficiency.