Bitcoin prediction has become an essential aspect of financial decision-making, driven by the cryptocurrency’s significant market volatility and potential for high returns. This work introduces a hybrid model that combines LSTM and GRU models to forecast Bitcoin prices. Traditional time series models like Linear Regression, FBProphet and ARIMAX often fail to capture the non-linear patterns of Bitcoin prices. Our proposed hybrid model addresses this by leveraging the strengths of both GRU and LSTM networks, optimized through a grid search approach for hyperparameter tuning. The model was evaluated against standalone GRU, LSTM, and traditional models. Our results demonstrate that the hybrid model significantly outperforms these models, achieving a MSE of 100.74, RMSE of 10.03, and R-squared of 0.98. Performance underscores the ability of the hybrid model to capture long-term dependencies and process them efficiently, providing a more accurate prediction of Bitcoin prices. The work shows the capability of combining Deep Learning techniques for financial time series forecasting.