This investigation evaluated the effect of various curing modes on the extent of microleakage of differently resin-based posterior composites. Box-form Class II (MO/OD) cavities (n=8) were prepared on proximal sides of 36 human sound mandibular molars. The cavities were restored by posterior composites (Filtek P90, Filtek Z250, and Filtek P60). The composite was placed in horizontal layers and irradiated occlusally by Astralis 10 light source at three modes of continuous intensities (conventional, high intensity power, and pulse-soft start). The specimens were thermocycled for 500 cycles at 5-55 degrees C, then immersed in 2% methylene blue solution for 4-h at 37 degrees C. The microleakage was examined by a stereomicroscope and scored occlusally and gingivally according to the ordinal scoring criteria. The Kruskal-Wallis revealed high significant differences between microleakage medians of the posterior composites examined (p<0.05). However, no significant differences were noted between microleakage medians at different modes of curing intensities occlusally (p=0.076) and there were significant differences gingivally (p=0.015). Mann-Whitney U-test showed a significant difference in microleakage for cavities restored by Z250 between high power and pulse modes of cure (p=0.006). The highest microleakage score was identified in the cavities restored with P60, while the lowest microleakage score was found in cavities restored by P90 specifically at conventional and pulse mode of cure. The light intensity modes have no significant effect on the microleakage, while the difference in composition of posterior resin composites investigated was the main factor for such a significant difference.