This QSAR model predicts the Log10(1/LC50) of organic chemicals and pesticides in Pimephales Promelas. The model was trained on data published by Netzeva et al. 2004. The dataset comprises 568 chemicals and their reported Log10(1/LC50) values. The model uses a Random Forest regressor trained on 80% of the dataset, while the remaining data were kept as a test set. During the training process, 10-fold cross-validation was performed. Four different methods (Leverage, Mahalanobis, Bounding Box, and City block) have been used to define the Domain of Applicability of the model. The model was evaluated in terms of R² in training and test datasets, and the mean R² from 10-fold cross-validation has been reported in the Metrics tab.