Abstract Bitcoin, as the pioneering cryptocurrency, has witnessed significant fluctuations in its price since its inception. Predicting these fluctuations is of great interest to investors, traders, and researchers alike. This paper presents a comparative study of Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) models for predicting Bitcoin prices. Cryptocurrency markets, particularly Bitcoin, exhibit high volatility and non-linearity, making accurate price prediction challenging. Recurrent Neural Networks (RNNs), including LSTM and GRU, have shown promise in capturing temporal dependencies and patterns in sequential data. In this study, we train and evaluate LSTM and GRU models separately on historical Bitcoin price data to forecast future price movements. Through comprehensive experimentation and analysis, we assess the performance of both models in terms of accuracy and effectiveness in Bitcoin price prediction. The performance of these models is evaluated based on various metrics such as Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The results demonstrate the effectiveness of LSTM and GRU in predicting Bitcoin prices and offer insights into their comparative performance.