Accurate forecasting of energy demand is crucial for effective management in Gujarat, India. This study compares the performance of SARIMA, Facebook (Fb) Prophet, and the Central Electricity Authority (CEA) models in predicting both total and peak energy demand using monthly data from April 2009 to March 2024. For total energy demand, SARIMA exhibited superior accuracy with a Root Mean Square Error (RMSE) of 1270.78 MU, a Mean Absolute Error (MAE) of 1024 MU, and a Mean Absolute Percentage Error (MAPE) of 7.01%, achieving a Structural Similarity Index Measure (SSIM) of 0.53. In contrast, the CEA model had higher RMSE and MAE values (1598.88 and 1313.41 MU, respectively), while Fb Prophet showed the highest RMSE of 2337.18 MU and a MAPE of 14.7%, despite a marginally better SSIM of 0.54. For peak energy demand, the CEA model performed best with the lowest RMSE of 1973.51 MW and a MAPE of 11.5%, while Fb Prophet and SARIMA had higher RMSEs and MAPEs, with Fb Prophet being more accurate than SARIMA. Fb Prophet projects a Compound Annual Growth Rate (CAGR) of 2.79% for total energy demand and 3.43% for peak energy demand from 2024 to 2029. These forecasts, covering the period from 2024 to 2029, provide crucial insights for policymakers in Gujarat to enhance energy planning and ensure supply reliability.