The acclimation, foraging behavior, and pollination efficiency of stingless bees of the species Nannotrigona perilampoides Cresson were evaluated in tomato (Lycopersicon esculentum Mill.) plants cultivated in two greenhouses. The greenhouses were divided into three areas of 16 m², and one of the following treatments was used for pollination: stingless bees (SB), mechanical vibration (MV), and no pollination (NP). Observations were conducted once a week from 0800 to 1600 hours during 2 mo. The acclimation of the bees to the greenhouses was estimated by the number of bees that did not return to the hive (lost bees) and by comparing the population of the colonies (brood and adults). The foraging activity of the bees across the day was evaluated by comparing the number of foragers per hour. The influence of environmental variables on the foraging activity was also analyzed. The pollination efficiency was compared among treatments through the percentage of fruit set, weight of individual fruit, kilograms of fruit produced per square meter, and the number of seed per fruit. The bees started foraging on the flowers ≈7 d after the colonies were introduced to the greenhouse. There was a decline in the population of the colonies across the experiment, but colonies did not die out. Correlations of environmental variables with the foraging activity of the bees showed that none of them had a significant influence on pollen foraging. However, water collection was positively correlated with the temperature and negatively correlated with the humidity inside the greenhouse. The estimation of the pollination efficiency per treatment showed that there were significant differences in fruit set in SB (83 ± 4.2) and MV (78.5 ± 6.4) compared with NP (52.6 ± 7.6). However, the average weight of the fruit was similar for the three treatments (65 g). There were significant differences for seed number in SB (200 ± 15.3) and MV (232 ± 21.4) compared with NP (120 ± 16.6). The productivity in kilograms of fruit per square meter was higher in SB (5.72 ± 0.61) and MV (5.66 ± 0.58 kg) compared with NP (3.34 ± 0.72). The number of seed was positively correlated with the weight of the fruit. We conclude that the use of Nannotrigona testaceicornis Rondani, for pollinating greenhouse tomatoes in tropical climates, could be an alternative to the use of highly defensive African-derived Apis mellifera or non-native bumble bees (Bombus spp.). However, more research is needed to evaluate the cost/benefit on large-scale greenhouse pollination using N. perilampoides Cresson against other bee species and pollination methods.