The aim of this study was to evaluate microleakage around class V restorations using a flowable composite compared to a hybrid composite. Forty class V cavities were prepared on buccal and lingual surfaces of 20 human teeth, with occlusal and cervical margins at the enamel and cementum/dentin levels, respectively. Specimens were divided into 2 groups with 10 samples each. Group 1: buccal cavities received Paama 2 (conventional bonding agent) + Wave (flowable composite); lingual cavities were restored with Paama 2 + Glacier (hybrid composite). Group 2: buccal cavities received Optibond Solo (self-priming bonding agent) + Wave; lingual cavities were restored with Optibond Solo + Glacier. After being stored in distilled water and finished, the teeth were thermocycled, immersed in a 50% silver nitrate solution and embedded in resin. They were sectioned and the depth of tracer penetration was scored. The results were analyzed using Kruskal-Wallis and Wilcoxon tests. The restorations with flowable composite and those with hybrid composite from the same group showed similar results of microleakage for both occlusal and cervical margins. Optibond Solo improved the sealing of the restorations when compared with Paama 2 (p < 0.01). None of the restorative materials completely sealed the tooth/restoration interface at the cervical margins.