In addition to the characteristics of soft actuators, such as flexibility and elasticity, biohybrid actuators also exhibit few distinctive functions, such as self-growth and self-healing. The previous research on biohybrid actuators has focused on culturing muscle cells and assembling them into micro-robots. These technologies are well-developed, however they lack the design and control methods found in the existing actuators that enable an appropriate performance. Therefore, we propose a simple muscle contraction model against external electrical stimulation, applying the model-based design and realizing the control of biohybrid actuators. The model comprises three sub-models-the electrical dynamic, physiological, and mechanical dynamic characteristics. The input of the model is the time-series stimulation voltage, and therefore, it can be applied to any complex stimulation waveform. The model parameters were identified with multiple square waves with different frequencies and amplitudes, based on the actual skeletal muscle of a toad. Subsequently, in the validation test, the actual muscle contraction was compared with the simulated force that was calculated using the identified parameters. Although the stimulations for the validation were different from those for the identification, the results obtained based on the model showed a good agreement with the experimental results. In addition, the optimal stimulation signal could be also calculated based on the model, to obtain the maximum net work. The findings of this study facilitate the development of model-based design and control methods for biohybrid actuators in the future, that will lead to significant development in this field.