This study is aimed to investigate the effect of multiple intelligences approach through the problem-based learning model (PBL) on the physics problem-solving abilities (PSA) of the tenth grade students. This study is an experimental research with a non-equivalent control group design. The sample was obtained through a purposive sampling technique, so the X-science 1 class was chosen as the experimental class and X-science 3 class as the control group. The experimental class was treated with a multiple intelligences approach through the PBL model, while the control class was given a conventional learning. Before being treated, the experimental class sample was given a multiple intelligence talent test questionnaire to find out the dominant intelligence in the class, after which the two samples were given a pre-test to find out their initial abilities. The results of the initial PSA test obtained an average experimental class value of 11.61 and a control class of 9.23. The homogeneity test results of the initial tests of the two samples had the same initial ability. The results of the PSA final test obtained were an average experimental class value of 72.32 and a control class of 60.62. The research hypothesis was tested by polled variance t-test with a significance level of 5%. Based on the results of the analysis, the value of tcount is greater than ttable, that is 11.9 > 1.99, so it can be concluded that there is an effect of multiple intelligences approach through PBL models on students’ physics problem solving.