This paper addresses the critical issue of electricity demand forecasting in Peru, where a significant number of electricity companies lack effective demand projection processes. The absence of accurate forecasting leads to increased operational costs, infrastructure investments, and customer dissatisfaction due to service interruptions. This study benchmarks various time series models, including ARIMA, Prophet, and Long Short-Term Memory (LSTM) networks, utilizing real demand data from the province of Concepción over the period from 2019 to 2023. We systematically analyze historical consumption data, employing methods such as Artificial Neural Networks (ANN), Long Short-Term Memory (LSTM), Support Vector Machine (SVM), and Autoregressive Integrated Moving Average (ARIMA). Model performance is evaluated using metrics like Mean Absolute Error (MAE) and Root Mean Square Error (RMSE), with results indicating that the LSTM model significantly outperforms others, achieving the lowest MAE (121,000) and RMSE (206,000). In contrast, the ANN model exhibited the poorest performance, while SVM and ARIMA showed intermediate results. This study underscores the importance of hyperparameter tuning and model selection in enhancing forecasting accuracy, ultimately aiming to improve operational efficiency and financial stability within the energy sector in Peru.