In its most general form, the mechanical constitutive behavior of any simple material connects the current stress state with its deformation history. The identification of such a relation and its material parameters is achieved through specific experimental setups, such as uniaxial tension-compression, biaxial tension-compression, and simple shear. Given some experimental data, calibration is usually accomplished through a method of least squares [1]. The associated definition of a loss function can quickly become ambiguous when combining different multi-axial experiments. Additionally, the method of least squares is by definition deterministic and therefore incapable of propagating uncertainty. At the same time, the need to assess the credibility of computational models continues to increase [2]. A Bayesian framework for calibrating constitutive models helps to overcome these challenges [3]. We construct a parameterization-invariant posterior to quantify the likelihood that a specific set of parameters corresponds to the expected behavior of a given simple material. We illustrate the approach using a real data set consisting of several different experiments performed on porcine stomach tissue for a rate-dependent constitutive model at finite strains. REFERENCES [1] Destrade M., Saccomandi G., Sgura I. Methodical fitting for mathematical models of rubber-like materials. Proc R Soc A 2017;473:20160811. [2] FDA Center for Devices and Radiological Health. Assessing the Credibility of Computational Modeling and Simulation in Medical Device Submissions. FDA, Rockville, USA; 2021. [3] Ranftl S., Müller T.S., Winderberger U., Brenn G., von der Linden W. A Bayesian approach to blood rheological uncertainties in aortic hemodynamics. Int J Numer Meth Biomed Eng 2022;e3576.