COVID-19 in Indonesia was first confirmed on March 2, 2020. On May 8, 2021, according Pusicov, Bandung City had reached 707 active cases. The increase in the number of active cases indicates whether the government can’t make good decisions in order to decrease the number of new cases or the service of hospital is not good in dealing with COVID-19 patients so the number of recovered is not likely increase. Therefore, forecasting the number of active COVID-19 cases in Bandung can be used to evaluate the government and hospital in facing COVID-19. To overcome the problem of analysis using conventional methods, this research will use Long Short Term Memory (LSTM) method. This method doesn’t require parametric assumptions and can be used for data with long time periods, as there is a cell state to overcome vanishing gradients in the Recurrent Neural Network method. Data is the number of positive cases, recovered, and death in 27 cities or regencies in West Java in the period March 02, 2020 – May 08, 2021 which were obtained from the pusicov and pikobar. Based on the results, the Mean Absolute Percentage Error (MAPE) is 31,46% for data testing.