Accurate prediction of monthly river discharge is essential for effective water resource management, flood control, and environmental planning. In this study, monthly river discharge prediction at Adityapur station in the Subarnarekha River basin is conducted using Support Vector Regression (SVR) in conjunction with Bayesian optimization to optimize SVR model hyperparameters, and the resulting model is compared with a simpler SVR variant. The model has been developed using ten years of monthly discharge data to simulate the period spanning from 2009 to 2019. Model performance is assessed and compared using evaluation metrics including Root Mean Square Error (RMSE), R-squared (R2), and Nash–Sutcliffe Efficiency (NSE). The Results show that the accuracy of the Bayesian optimization-based SVR model is high and better than the SVR model. The R2 values for Bayesian optimization-based SVR and SVR models are 0.9358 and 0.8859 respectively. Similarly, The RMSE values for Bayesian optimization-based SVR and SVR models are 50.106 and 66.84 m3/s and NSE values are 0.9238 and 0.7772 respectively. Based on the results, it is concluded that the model’s efficiency can be increased by optimizing the SVR model’s hyperparameters using Bayesian optimization.