A generalized height–diameter (h-d) model was developed in order to predict the total height of individual trees in uneven-aged stands in the region of El Salto, Durango (Mexico). Seven generalized h-d equations were evaluated and the equation proposed by Sharma and Parton, which includes the diameter at breast height of the tree and the quadratic mean diameter and the dominant height of the stand as independent variables, was selected as the best model. In order to address the among-plot variability, a non-linear mixed-effects modelling approach was used to fit the selected model for all the species or groups of species. The mixed model included a random parameter that affected the model asymptote. Calibration of the obtained h-d model for a particular species or group of species in a plot of interest was carried out with only a single randomly selected tree from the species or group of species within the plot. The stochastic component added to the mixed-effects model enabled the observed natural variability in heights within diameter classes for the same stand to be mimicked, thereby providing more realistic predictions. The equation developed represents a new tool for evaluation and management of uneven-aged stands in the region.