ABSTRACT.- Apple pedestrian system allows performing all cultural practices from the ground and allows the mechanization of some practices such as thinning and pruning. The objective of this study was to evaluate the performance of different training systems in a pedestrian orchard. A trial with ?Gala Baigent? (Brookfield®) apples was installed at INIA Las Brujas, Canelones, Uruguay, in 2014/2015 growing season. Five training systems were evaluated:one leader/spindle, three and four leaders, guyot (around 5 leaders), and horizontal palmette. Trees were grafted onto ?M.9 NAKB T337? and spaced 3 m between rows and 0.75 m (one leader), 1 m (three leaders), and 1.5 m (four leaders, guyot, and horizontal palmette) within the row. Trees were laterally pruned at 0.25 m from the trunk on each side and topped at 2.20 m after harvest with a heading machine. Significant differences in number of fruit and yield per tree were observed, where guyot and horizontal palmette had greater values as a direct effect of the space covered by the tree in the fruiting wall. Estimated cumulative yield of different training systems was between 50 and 70 t ha-1, considering the first three harvest (2017-2019). Although one leader was the most productive training system in the first production year in MURBA, those with multi-leader trees showed a better vegetative balance and less hand labor requirement. After the first harvest, different training systems completely covered the fruiting wall; no differences were observed in yield per hectare between the evaluated treatments. Fruit quality attributes were not affected. Time of winter pruning was not affected; however, significant differences in pruned fresh weight have been identified and negatively correlated with number of leaders per hectare over the seasons. Even though this system requires an economic analysis, the different training systems in MURBA are productive and suitable for handling pedestrian apple crops in Uruguayan conditions. © 2022 International Society for Horticultural Science. All rights reserved.