In initial observations, it was seen that students did not write down what they knew (interpretation), were not precise in making mathematical models (analysis), were less precise in using strategies to solve problems (evaluation), and had no conclusions (inference). After paying attention, students have different learning styles. However, the learning model used has not been able to improve mathematical critical thinking skills in terms of learning style. This research aims to find out how mathematical critical thinking skills are treated with the STEM-based STAD learning model in terms of learning styles. This type of research is quantitative research with research subjects namely class X students at MAN 1 Muaro Jambi. The research design used was a posttest only control design. The population was 115 students from 4 classes and the sample consisted of 3 classes, namely experimental class I (29 students), experimental class II (29 students), and control (28 students). The sampling technique was carried out using the cluster random sampling technique. The instruments in this research were observation sheets, learning style questionnaires, and mathematical critical thinking ability tests. The data obtained in this study were analyzed using a two-way ANOVA test. The results of the research show that: 1) There is an influence of the application of the STEM-based STAD model on mathematical critical thinking skills where the significance value of the learning model is , so is rejected or is accepted, 2) There is no influence of learning style on mathematical critical thinking skills with The significance value for learning style is , so is rejected or is accepted, and 3) There is no interaction between the application of the STEM-based STAD model and learning style on mathematical critical thinking skills with a significance value of , so is rejected or accepted. The results of the research concluded that the STEM-based STAD learning model applied in experimental class I was more effective for mathematical critical thinking skills, as evidenced by the average mathematical critical thinking ability of the STEM-based STAD learning model of which is greater than the average. using STAD and direct instruction.