ProTox-II is a webserver for predicting the toxicity of chemicals. It integrates molecular similarity, pharmacophores, fragment propensities, and machine learning models to predict various toxicity endpoints, including acute toxicity, hepatotoxicity, cytotoxicity, carcinogenicity, mutagenicity, immunotoxicity, and adverse outcomes pathways (Tox21). The models are built on data from in vitro and in vivo assays, and have been validated on independent external sets, showing strong performance. ProTox-II provides a freely available webserver for in silico toxicity prediction, accessible at http://tox.charite.de/protox_II. The webserver takes a two-dimensional chemical structure as input and reports the possible toxicity profile for 33 models with confidence scores, an overall toxicity radar chart, and three most similar compounds with known acute toxicity. The ProTox-II platform includes five classification steps: acute toxicity, organ toxicity, toxicological endpoints, toxicological pathways, and toxicity targets. The platform uses various models, including machine learning algorithms like Random Forest and Support Vector Machine, and incorporates molecular fingerprints for prediction. The models are validated using cross-validation and external validation, with performance measures such as balanced accuracy, AUC–ROC, and kappa values. The platform also provides application cases, such as the prediction of Etonogestrel and Tolcapone, demonstrating its ability to predict toxicity endpoints and identify potential toxicological pathways and targets. ProTox-II is a freely available computational toxicity prediction webserver that enables the prediction of the largest number of toxicity endpoints. It provides insights into the possible molecular mechanisms behind toxic responses and supports risk assessments for regulatory decisions. Future updates will focus on improving the characterization of clinically relevant adverse effects and considering species and inter-individual genetic differences. Regular updates and new data will be added to maintain the high standard of the platform.ProTox-II is a webserver for predicting the toxicity of chemicals. It integrates molecular similarity, pharmacophores, fragment propensities, and machine learning models to predict various toxicity endpoints, including acute toxicity, hepatotoxicity, cytotoxicity, carcinogenicity, mutagenicity, immunotoxicity, and adverse outcomes pathways (Tox21). The models are built on data from in vitro and in vivo assays, and have been validated on independent external sets, showing strong performance. ProTox-II provides a freely available webserver for in silico toxicity prediction, accessible at http://tox.charite.de/protox_II. The webserver takes a two-dimensional chemical structure as input and reports the possible toxicity profile for 33 models with confidence scores, an overall toxicity radar chart, and three most similar compounds with known acute toxicity. The ProTox-II platform includes five classification steps: acute toxicity, organ toxicity, toxicological endpoints, toxicological pathways, and toxicity targets. The platform uses various models, including machine learning algorithms like Random Forest and Support Vector Machine, and incorporates molecular fingerprints for prediction. The models are validated using cross-validation and external validation, with performance measures such as balanced accuracy, AUC–ROC, and kappa values. The platform also provides application cases, such as the prediction of Etonogestrel and Tolcapone, demonstrating its ability to predict toxicity endpoints and identify potential toxicological pathways and targets. ProTox-II is a freely available computational toxicity prediction webserver that enables the prediction of the largest number of toxicity endpoints. It provides insights into the possible molecular mechanisms behind toxic responses and supports risk assessments for regulatory decisions. Future updates will focus on improving the characterization of clinically relevant adverse effects and considering species and inter-individual genetic differences. Regular updates and new data will be added to maintain the high standard of the platform.