Wind turbines are inherently intermittent energy sources that predominantlydepend on meteorological conditions. Accurate prediction of wind turbine active powergeneration is crucial for optimizing the operation of modern power systems rich in renewableenergy sources, improving system stability, enhancing the efficiency of production andconsumption balancing, optimizing the electricity market, and numerous other aspects. Thedevelopment and application of advanced artificial intelligence and machine learning (ML)methods can further increase prediction accuracy, enabling the system to better adapt to thevariable nature of wind energy. This paper investigates the efficiency of two machine learningapproaches for predicting wind turbine active power, focusing on ML models based onrecurrent neural networks (LSTM and GRU) and Transformer models adapted for processingand predicting time series data. Transformer models have recently gained prominence in timeseries analysis due to their ability to effectively recognize long-term dependencies in datathrough the self-attention mechanism, enabling parallel sequence processing and moreaccurate identification of relevant patterns in the data. The choice of transformer modelarchitecture and parameter tuning is conditioned by two iterative processes. The validation ofthe developed ML model is performed using a practical open-source dataset that contains alarge number of input features related to meteorological and operational system parameters.The selection of the most relevant features is based on their correlation with the targetvariable (in this case, active power), reducing the dimensionality of the problem andimproving model efficiency. The experimental evaluation includes an analysis of modelperformance using standard metrics (RMSE and MAEI), while simultaneously examining thetransformer architecture and its parameter set. The obtained results provide insight into theadvantages and limitations of both approaches in various scenarios of short-term and long-term prediction, with the aim of improving broader prediction strategies and optimizing windfarm operations.