OBJECTIVE Pelvic floor muscles (PFM) are deleteriously affected by vaginal birth, which contributes to the development of pelvic floor disorders. To mechanistically link these events, experiments using animal models are required, as access to human PFM tissue is challenging. In choosing an animal model, a comparative study of PFM design is necessary, since gross anatomy alone is insufficient to guide the selection. METHODS Human PFM architecture was measured using micromechanical dissection and then compared with mouse (n = 10), rat (n = 10), and rabbit (n = 10) using the Architectural Difference Index (ADI) (parameterizing a combined measure of sarcomere length-to-optimal-sarcomere ratio, fiber-to-muscle-length ratio, and fraction of total PFM mass and physiological cross-sectional area (PCSA) contributed by each muscle). Coccygeus (C), iliocaudalis (IC), and pubocaudalis (PC) were harvested and subjected to architectural measurements. Parameters within species were compared using repeated measures analysis of variance (ANOVA) with post hoc Tukey's tests. The scaling relationships of PFM across species were quantified using least-squares regression of log-10-transformed variables. RESULTS Based on the ADI, rat was found to be the most similar to humans (ADI = 2.5), followed by mouse (ADI = 3.3). When animals' body mass was regressed against muscle mass, muscle length, fiber length, and PCSA scaling coefficients showed a negative allometric relationship or smaller increase than predicted by geometric scaling. CONCLUSIONS In terms of muscle design among commonly used laboratory animals, rat best approximates the human PFM, followed by mouse. Negative allometric scaling of PFM architectural parameters is likely due to the multifaceted function of these muscles.