Soft robotics is emerging as a compelling alternative to rigid robotics for artificial heart devices, offering enhanced adaptability and compliance through the use of elastomers, silicone, or fabrics, while relying on comparatively simple actuation mechanisms. In particular, fabric-based soft robotic hearts have gained attention due to their low weight, biocompatibility, and tailorable force transmission enabled by anisotropic reinforcements. Despite these advances, a comprehensive framework to study intrinsic device mechanics, identify durability-limiting features, and rationally guide design modifications is still lacking, as these tasks remain largely inaccessible to purely experimental approaches. In this work, we present a computational model of the Less In, More Out (LIMO) soft artificial heart. The device consists of flat fabric pouches that are fluidically actuated into approximately cylindrical shapes, thereby reducing the internal volume and ejecting fluid. Our model resolves the dynamic, nonlinear deformation of the device using a finite-element formulation. We benchmark the model against experimental pressure–volume measurements across a range of pouch numbers and afterload conditions. Consistent with experiments, we find that devices with fewer pouches exhibit lower input–output ratios and reduced mechanical efficiency compared to devices with more pouches, and that mechanical efficiency increases with afterload. The simulations reveal stress concentrations and low-fatigue-life regions along heat-sealed seams and in buckling-prone areas, directly corresponding to experimentally observed failure modes and providing a durability metric inaccessible through experiments alone. We further demonstrate the utility of the framework for in silico design exploration: targeted modifications to valve support geometry and spatial variations in fabric compliance reduce peak stresses by approximately 10% while preserving cardiac output and efficiency. Finally, we investigate the behavior of the soft ventricle under physiologically relevant hemodynamic conditions by coupling the finite-element model to a lumped-parameter circulation model and an unsteady flow solver. This coupled framework allows us to assess the influence of hemodynamic loading on device efficiency and intraventricular flow dynamics.