This study aims to determine whether the TPS or GI learning models are more effective for students' understanding of mathematical concepts. Conducted at Dr. Wahidin Mlati Junior High School during the academic year 2023/2024, the research utilized a quasi-experimental design with a non-equivalent posttest only control group and simple random sampling technique. In experimental class 1, a significance value of 0.202 > 0.05 was obtained for normality test, and in experimental class 2, a significance value of 0.155 > 0.05 was obtained, indicating normal distribution of both classes' posttest scores. The homogeneity test of posttests yielded a significance value of 0.09 > 0.05, indicating homogeneity of variance between the two classes. Regarding the hypotheses testing, the first hypothesis showed with 11.68 > 1.69, indicating that the TPS learning model effectively helps students understand mathematical concepts. The second hypothesis showed with 4.92 > 1.69, indicating that the GI learning model effectively influences students' ability to understand mathematical concepts. The third hypothesis showed with 2.87 > 1.99, indicating that the TPS learning model is more effective than the GI learning model for understanding mathematical concepts among seventh grade students at Dr. Wahidin Mlati Junior High School