Rock salt constitutive models are used to simulate the evolution of mines, boreholes, storage caverns for gases and liquids, and nuclear waste repositories in rock salt formations. A wide variety of thermo-mechanical constitutive models have been proposed for rock salt, yet even the damage-free (micro-crack-free) thermoviscoplastic behavior remains difficult to capture. The Munson-Dawson model (Reedlunn et al. 2022), for example, can be calibrated against damage-free constant stress tests with low to medium steady-state strain rates (10–12 to 10–81/s), but such a calibration fails to represent damage-free constant strain rate tests at high strain rates (10–6 to 10–4 1/s). Capturing the damage-free behavior at these high strain rates is important because high strain rates are frequently used to characterize the mechanical behavior of damaged salt (dilated salt with micro-cracks). A constitutive model must first adequately capture damage-free behavior before attempting to represent how damage degrades salt’s strength. This paper presents a new model for the damage-free behavior of salt. Model development was influenced by a variety of models for metals (see J. Chaboche (2008) for a review) and the Aubertin, Yahya, et al. (1999) model for salt. Section 2 defines the new model formulation, Section 3 briefly discusses the formulation, Section 4 details four model calibrations, Section 5 partially validates the model, and Section 6 provides a short summary. Several preliminaries bear mentioning before defining the model formulation. First, compressive strains and stresses are treated as positive. Second, variables represented by capital letters are material constants, while variables represented by lower case letters are functions of other variables. Third, this section presents the model in an infinitesimal strain setting, but one can easily have extended the model into the finite deformation realm using hypoelasticity.