Students today must have 21st-century skills to succeed in the workplace. Unfortunately, some pupils lack 21st-century skills due to an inefficient educational system. The study's overarching goal was to compare and contrast the problem-solving abilities of students exposed to Problem-Based Learning and those exposed to Discovery Learning, focusing on the relationship between the two in terms of students' self-efficacy. The research design used a quasi-experimental non-equivalent control group design. The population in this study was fourth-grade elementary school students in the Citapen area, Bogor Regency, and the sampling technique was simple random sampling at a state elementary school in Bogor Regency. The research sample was 96 students consisting of 2 experimental and one control class. The research instrument is a problem-solving ability test and a self-efficacy questionnaire. The restudy's results were tested using two-way ANOVA, showing differences in students' problem-solving abilities with high, medium, and low self-efficacy taught by Problem-Based Learning and Discovery Learning. It is hoped that the results of this study can provide a reference as a learning alternative that could improve students' problem-solving abilities.