When prior information is in the form of marginal distributions, it is advantageous to consider families of bivariate distributions with specified marginals when modelling bivariate data. Among these families of bivariate distributions, the Farlie-Gumbel-Morgenstern (FGM) family was studied extensively by many authors. Several modifications to the FGM family have been proposed in the literature to increase the range of the correlation between its marginals. One important limitation of the FGM family is that the correlation coefficient between its marginals is restricted to a narrow range $\left[-\frac{1}{3},\frac{1}{3}\right].$ One of the most pliable and robust extensions of the classical FGM family of bivariate distributions is the Sarmanov family, which was proposed and used by Sarmanov (1974) as a new model of hydrological processes, inter alia. Despite the salient and almost unique features of this family, it is never used in the literature. We study the concomitants of $k-$record values based on Sarmanov exponential distribution. Also, we derived the joint distribution of concomitants of $k-$record values based on this family.