The age of a star cannot be measured but only inferred from observational quantities related to age. A wellknown stellar age greatly impacts our understanding of evolution on much larger scales, such as exoplanetary systems or the history of the Milky Way. For stars in the red-giant phase, comparison of observations with evolutionary models delivers ages that are typically determined with statistical uncertainties between 30 and 50%. These uncertainties originate from ill-constrained intricacies of the stellar input physics and related parameter degeneracies that cannot be resolved from modeling single stars. Double-lined spectroscopic (SB2) binaries, where certain spectroscopic lines are available for both components, provide a unique opportunity to constrain stellar physics, potentially leading to better stellar ages. Additionally, detectable oscillation-power excesses in one or both components in a binary system enable an independent calculation and confirmation of stellar properties, such as mass, radius, and composition. In this talk, we present the in-depth analysis of KIC9163796, a double-lined oscillating red-giant & red-giant binary system. Although having a mass ratio of almost unity, both components vary significantly in effective temperature, luminosity, radius, and lithium abundance. We use the differences observed in the stars to constrain the two combined models for the primary and secondary. By creating a multi-dimensional model grid for the combined modeling approach of both components, we used the stellar evolution code MESA and the theoretical frequency spectrum of the stellar oscillation modes to evaluate the best values of the masses, the convective parameters, and initial helium, at the constraints of identical ages and metallicity. This modeling approach also allowed us to determine the age of the system with a 20 % precision (tau = 2.5 ± 0.5 Gyr). In conclusion, this binary system provides an important benchmark for improved age determination on the red-giant branch.