The depletion of tropical rainforests is on the increase and plantation forests have been identified as a quick-fix measure. Sustainable management of plantation forests for optimum productivity is, therefore, essential. Hence, this study was carried out to develop site management models (site index) for Tectona grandis plantations in Gambari, Oluwa, Omo, and Sankpoba Forest Reserves, Nigeria. The data for the study were collected from six (6) age series (6, 8, 10, 12, 14, and 16 years old) in each location. Five (5) sample plots (25m x 25m) were randomly selected per age series (making a total of 120 sample plots) and two-thirds of the data were used for calibration, while the remaining one-third was used for validation purpose. Based on the index age of 10 years, the selected site index models constructed using the regression model approach in Gambari, Oluwa, Omo, and Sankpoba were SI = exp [ln(Hd) – 8.65 (A^-1 – 0.1)], SI = exp [ln(Hd) – 5.77 (A^-1 – 0.1)], SI = exp [ln(Hd) – 9.31 (A^-1 – 0.1)], and SI = exp [ln(Hd) – 4.35 (A^-1 – 0.1)], respectively. Using the graphical approach and proportional curve method, the plantations were stratified into five site quality classes, I to V (I = the best site and V = poor site). A 10-year-old Teak stand from this study attained average dominant heights (m) of 16.2, 14.1, and 9.8 (Gambari); 19.5, 16.0, and 12.3 (Oluwa); 21.8, 16.2, and 10.5 (Omo), and 15.5, 11.4, and 8.9 (Sankpoba) on site classes I, III, and V, respectively. This implies that a stand with a height development pattern similar to curve I is growing on the best site and has a higher volume if production than a stand with a height development pattern similar to curve V. It is recommended that site classes I to III should be used for Teak plantations in this study areas