In quadrupeds, the cervical and lumbar circuits work together to achieve the speed-dependent gait expression. While most studies have focused on how local lumbar circuits regulate limb coordination and gaits, relatively few studies are known about cervical circuits and even less about locomotor gaits. We use the previously published models by Danner et al. (DANNER, S. M., SHEVTSOVA, N. A., FRIGON, A., and RYBAK, I. A. Computational modeling of spinal circuits controlling limb coordination and gaits in quadrupeds. eLife, 6, e31050 (2017)) as a basis, and modify it by proposing an asymmetric organization of cervical and lumbar circuits. First, the model reproduces the typical speed-dependent gait expression in mice and more biologically appropriate locomotor parameters, including the gallop gait, locomotor frequencies, and limb coordination of the forelimbs. Then, the model replicates the locomotor features regulated by the M-current. The walk frequency increases with the M-current without affecting the interlimb coordination or gaits. Furthermore, the model reveals the interaction mechanism between the brainstem drive and ionic currents in regulating quadrupedal locomotion. Finally, the model demonstrates the dynamical properties of locomotor gaits. Trot and bound are identified as attractor gaits, walk as a semi-attractor gait, and gallop as a transitional gait, with predictable transitions between these gaits. The model suggests that cervical-lumbar circuits are asymmetrically recruited during quadrupedal locomotion, thereby providing new insights into the neural control of speed-dependent gait expression.