Various modifications to standard "rigid" anterior cervical plate designs (constrained plate) have been developed that allow for some degree of axial translation and/or rotation of the plate (semi-constrained plate)-theoretically promoting proper load sharing with the graft and improved fusion rates. However, previous studies about rigid and dynamic plates have not examined the influence of simulated muscle loading. The objective of this study was to compare rigid, translating, and rotating plates for single-level corpectomy procedures using a robot testing system with follower load. In-vitro biomechanical test. N=15 fresh-frozen human (C3-7) cervical specimens were biomechanically tested. The follower load was applied to the specimens at the neutral position from 0 to 100N. Specimens were randomized into a rigid plate group, a translating plate group and a rotating plate group and then tested in flexion, extension, lateral bending and axial rotation to a pure moment target of 2.0Nm under 100N of follower load. Range of motion, load sharing, and adjacent level effects were analyzed using a repeated measures analysis of variance (ANOVA). No significant differences were observed between the translating plate and the rigid plate on load sharing at neutral position and C4-6 ROM, but the translating plate was able to maintain load through the graft at a desired level during flexion. The rotating plate shared less load than rigid and translating plates in the neutral position, but cannot maintain the graft load during flexion. This study demonstrated that, in the presence of simulated muscle loading (follower load), the translating plate demonstrated superior performance for load sharing compared to the rigid and rotating plates.