Diameter at the breast and height of trees are the most important components in the forest inventory. Measuring the diameter of trees is simpler and less costly than height; hence, some equations are used to predict height by diameter of trees. In the present study, the ability to use different diameter-height models for estimating the height of beech trees (Fagus oriantalis Lipsky) in uneven age and mixed stands in Hyrcanian forest, North of Iran. In this study, a systematic-randomly sampling method with 150 × 200m network (0.1 ha) was used. Diameter and height of the thickest and nearest trees (690 individuals) to the center of circular sample plots (345) was measured. 70% of the data was used for modeling and the remaining 30% was used for evaluating estimated models. Using 20 nonlinear regression models including 11 two-parameter models and 9 three-parameter models, the relationship between height as a dependent variable and diameter as an independent variable was considered and analyzed. In order to evaluate the models and select the best model, the validity of the statistical models was evaluated using RMSE and BIAS. The results of the model evaluation criteria did not differ significantly. Korf, Ratkowsky, Naslund and Weibull models with root mean square error of 4.17, 4.19, 4.21, and 4.23 and BIAS of 0.17, -0.38, -0.55 and -0.1, respectively had a good ability to accurately estimate the height of beech trees. According to the region conditions, these models can be used to estimate the height of beech trees in broadleaved and mixed forest of northern Iran.