Aims Mid -level constraint designs for total knee arthroplasty (TKA) are intended to reduce coronal plane laxity. Our aims were to compare kinematics and ligament forces of the Zimmer Biom-et Persona posterior-stabilized (PS) and mid -level designs in the coronal, sagittal, and axial planes under loads simulating clinical exams of the knee in a cadaver model. Methods We performed TKA on eight cadaveric knees and loaded them using a robotic manipulator. We tested both PS and mid -level designs under loads simulating clinical exams via applied varus and valgus moments, internal-external (IE) rotation moments, and anteroposterior forces at 0 degrees, 30 degrees, and 90 degrees of flexion. We measured the resulting tibiofemoral angulations and translations. We also quantified the forces carried by the medial and lateral collateral ligaments (MCL/LCL) via serial sectioning of these structures and use of the principle of superposition. Results Mid -level inserts reduced varus angulations compared to PS inserts by a median of 0.4 degrees, 0.9 degrees, and 1.5 degrees at 0 degrees, 30 degrees, and 90 degrees of flexion, respectively, and reduced valgus angulations by a median of 0.3 degrees, 1.0 degrees, and 1.2 degrees (p <= 0.027 for all comparisons). Mid -level inserts reduced net IE rotations by a median of 5.6 degrees, 14.7 degrees, and 17.5 degrees at 0 degrees, 30 degrees, and 90 degrees, respectively (p = 0.012). Mid -level inserts reduced anterior tibial translation only at 90 degrees of flexion by a median of 3.0 millimetres (p = 0.036). With an applied varus moment, the mid -level insert decreased LCL force compared to the PS insert at all three flexion angles that were tested (p <= 0.036). In contrast, with a valgus moment the mid -level insert did not reduce MCL force. With an applied internal rotation moment, the mid -level insert decreased LCL force at 30 degrees and 90 degrees by a median of 25.7 N and 31.7 N, respectively (p = 0.017 and p = 0.012). With an external ro-tation moment, the mid -level insert decreased MCL force at 30 degrees and 90 degrees by a median of 45.7 N and 20.0 N, respectively (p <= 0.017 for all comparisons). With an applied anterior load, MCL and LCL forces showed no differences between the two inserts at 30 degrees and 90 degrees of flexion. Conclusion The mid -level insert used in this study decreased coronal and axial plane laxities compared to the PS insert, but its stabilizing benefit in the sagittal plane was limited. Both mid -level and PS inserts depended on the MCL to resist anterior loads during a simulated clinical exam of anterior laxity.