Purpose: This research was conducted to investigate the potential of the ERCoRe learning model in empowering the retention of students' of different academic ability. Research Methods: This was a quasi-experimental research using pre-test and post-test non-equivalent control group design of 2x2. There were two independent variables. The first variable was the learning model consisting of the ERCoRe model and conventional learning, and the second variable was academic ability, consisting of upper and lower levels of academic ability. The dependent variable was the students' retention. The samples for this research were the students of class X in Pangkep District, Indonesia. The data from this research were analysed by using ANCOVA, followed by Least Significant Different (LSD). Findings: The ERCoRe learning model was shown to have more potential for improving the students' retention than conventional learning (11.58%). The interaction between the ERCoRe learning model and academic ability did not have an effect on students' retention, but it was seen from the combination groups that the retention of the higher academic ability students who experienced ERCoRe learning was higher (significantly different) than that of the other combination groups. Implications for Research and Practice: Teachers need to implement the ERCoRe learning model because this learning model can improve the level of students' retention.