Abstract We present a method to infer ocean diapycnal diffusivity based on high‐resolution ocean model predictions of the depth‐dependent viscous dissipation associated with internal wave shear. This method relies on recent advances in modeling and the parameterization of stratified turbulent mixing processes. Especially important in the latter regard is the distinction between irreversible and reversible mixing processes associated with internal wave breaking. Utilizing the Bouffard–Boegman (BB) compilation of data, we derive depth‐dependent profiles of diapycnal diffusivity from viscous dissipation rates obtained from downscaled internal wave fields of the global ocean simulation LLC4320. Our methodology displays some skill in matching observationally‐informed inferences of diapycnal diffusivity and demonstrates that the KPP‐based production of diapycnal diffusivity fails to account for the distinction between reversible and irreversible mixing components. This work provides a framework for further improving the parameterization of mixing processes in large scale climate models through simulations of the background internal wave field.