The primary objective of this study is to evaluate the effectiveness of selected machine learning architectures for the detection and spatial localization of leaks in water distribution networks using pressure telemetry data. The study compares five models: a convolutional neural network (CNN), a recurrent neural network (LSTM), and three graph-based architectures (GNN, DGT, and STGT). Experiments were conducted on a dataset comprising 35 topologically diverse water distribution networks, using normal-operation and synthetic leak scenarios generated with the EPANET simulator and the WNTR library. The models were evaluated in terms of scenario-level leak detection, node-level localization performance, spatial localization error expressed as topological hop-distance, and inference time. The graph-based architectures exhibited a pronounced over-prediction tendency: although they achieved very high node-level recall, they also generated a large number of false-positive node classifications, limiting their practical applicability in the investigated configuration. Among the evaluated architectures, the LSTM model provided the most favorable trade-off between localization sensitivity, false-positive rate, spatial precision, and computational efficiency. It achieved a node-level sensitivity of 93.44%, a node-level false-positive rate of 21.96%, a mean false-positive hop-distance of 1.26, and an inference time of approximately 2 seconds. The results indicate that LSTM-based architectures are promising candidates for further development of decision-support systems for water distribution network monitoring, while graph-based models require additional probability calibration, threshold optimization, and sparsity-oriented regularization.