Unicondylar arthroplasty was performed using robotic medial unicompartmental knee arthroplasty (R-mUKA) and gap-balancing instrumentation. Our hypothesis was that robotic unicondylar knee arthroplasty accurately restores component positioning and lower limb alignment when compared to preoperative planning with actual implantation throughout the range of knee motion due to proper knee balancing. Data were collected prospectively and were analysed for patients undergoing RM-UKA. A cemented UKA was implanted using the MAKO® robotic system. Lower limb alignment at 0°, 30°, 45°, 60° and 90° of flexion was recorded of the native knee, with the trial components in place and finally after component implantation. A spacer according to the femorotibial gap was introduced and the alignment was measured. The position of the final component was planned based on three-dimensional computed tomography images before making the bone cuts. The positioning of the femoral and tibial components was analysed in all three planes. A total of 52 patients were included (mean age 66.3 ± 6.7 years; 34 males, 18 females). The difference in femoral component position after planning and final implantation was 0.04° ± 0.58° more valgus in the coronal plane (p = 0.326) and 0.6° ± 1.4° more flexion relative to the sagittal plane (p = 0.034). The tibial component was placed in the coronal plane in 0.3° ± 0.8° of more varus (p = 0.113) and in the sagittal plane in 0.6° ± 1.2° of more posterior tibial slope (p = 0.001). Lower limb alignment of the native knee in extension was 5.8° ± 2.6° of varus and changed to 3° ± 2.1° varus after UKA (p ≤ 0.01). R-mUKA helps to achieve the target of alignment and component position without any significant differences to the planning. Ligament balancing causes non-significant changes in component position. It allows optimal component position even for off-the-shelf implants respecting the patient's specific anatomy. II.