Conclusions: Computational fluid dynamics (CFD) methods can provide detailed information on airflow characteristics in models of septal deviation. Different locations of septal deviation lead to variation of nasal airflow. Both the location of the septal deviation and the inferior turbinate hypertrophy in the concave side may play an important role in airflow patterns and airflow velocity. Objective: To investigate the airflow patterns and air velocity in different septal deviation models during inspiration, using CFD methods. Methods: Commercial software was used to construct three-dimensional (3D) models of nasal cavities with paranasal sinuses from the computed tomography (CT) scans of 15 patients with septal deviation and 4 controls. Considering the location of the most prominent point of the nasal septum, patients were classified into caudal, anterior, and media deviation groups. Results: Unlike airflow in the controls, airflow in the septal deviation models showed asymmetry in bilateral nasal cavities. The airflow patterns varied in the convex and concave sides in different septal deviation models. Caudal septal deviation models had the maximal peak velocity, while the the minimal peak velocity was found in the media deviation models. The peak velocity was not always located in the convex side, but was sometimes in the concave side.