This project bitcoin's price is forecated using the Machine Learning techniques, Artificial Neural Networks on. The dataset used for this is downloaded from kaggle as historical bitcoin data. The FFNN, DNN and RNN models are predicting the price very close to actual value. The accuracy is evaluted by errors, RMSE,MAE and MAPE. In ann-models, models are developed that are individually can be tested. fyp_web_app is the app developed in flask where are the graphs and tables are shown for these Algorithms.