ProTox-II is a webserver designed for predicting the toxicity of chemicals, incorporating molecular similarity, pharmacophores, fragment propensities, and machine-learning models. It predicts various toxicity endpoints, including acute toxicity, hepatotoxicity, cytotoxicity, carcinogenicity, mutagenicity, immunotoxicity, adverse outcome pathways (AOPs), and toxicity targets. The models are built using data from both in vitro and in vivo studies and have been validated on independent external datasets. ProTox-II provides a user-friendly interface for inputting chemical structures and outputs predictions with confidence scores, an overall toxicity radar chart, and the three most similar compounds with known acute toxicity. The platform includes 33 models and is freely available to toxicologists, regulatory agencies, and researchers. The article details the methods used for model development, validation, and application cases, highlighting the platform's performance and potential in supporting risk assessments and drug discovery. Future updates aim to enhance the platform's capabilities by adding new data and endpoints and considering species and inter-individual genetic differences.ProTox-II is a webserver designed for predicting the toxicity of chemicals, incorporating molecular similarity, pharmacophores, fragment propensities, and machine-learning models. It predicts various toxicity endpoints, including acute toxicity, hepatotoxicity, cytotoxicity, carcinogenicity, mutagenicity, immunotoxicity, adverse outcome pathways (AOPs), and toxicity targets. The models are built using data from both in vitro and in vivo studies and have been validated on independent external datasets. ProTox-II provides a user-friendly interface for inputting chemical structures and outputs predictions with confidence scores, an overall toxicity radar chart, and the three most similar compounds with known acute toxicity. The platform includes 33 models and is freely available to toxicologists, regulatory agencies, and researchers. The article details the methods used for model development, validation, and application cases, highlighting the platform's performance and potential in supporting risk assessments and drug discovery. Future updates aim to enhance the platform's capabilities by adding new data and endpoints and considering species and inter-individual genetic differences.