The butterfly algorithm is a promising method for efficiently solving N-body problems, offering potential advantages in accuracy, efficiency, and scalability. In this study, we comprehensively evaluate the performance of the butterfly algorithm in the context of N-body simulations, focusing on its accuracy, efficiency, and scalability. To assess accuracy, we compare the results of the butterfly algorithm for simple test cases, calculating the root mean square error (RMSE) as a prime metric. A low RMSE value means the algorithm can closely match the analytical solutions. This shows that it is good at accurately simulating gravitational interactions in the N-body system. Additionally, we investigate the algorithm’s efficiency by measuring the computational time required to simulate N-body systems of varying sizes. Our evaluation includes a comparison with a brute-force algorithm to highlight the efficiency gains provided by the butterfly algorithm. We analyze the scalability of the butterfly algorithm by increasing the number of particles in the simulations and measuring the corresponding computation time. Through these analyses, we aim to provide insights into the practical utility of the butterfly algorithm for large-scale N-body simulations. Our findings demonstrate the accuracy of the butterfly algorithm, with low RMSE values observed across different scenarios. Whereas our scalability analysis reveals the algorithm’s ability to handle increasing computational loads while maintaining reasonable execution times. Our study contributes to a better understanding of the butterfly algorithm’s capabilities and limitations, offering valuable insights for its application in various scientific and engineering domains requiring N-body simulations.