Low mathematics learning outcomes and students' mathematics activities and the learning process tend to be centered on the teacher. One alternative to overcome this problem is to apply the Auditory Intellectually Repetition (A.I.R.) learning model. Student activity is seen from the student activity sheet data, including whether the student is included in the criteria categorized as active or very active. The study aimed to determine whether the ability of students' mathematics learning outcomes and students' mathematics activities using the Auditory Intellectually Repetition (A.I.R.) model was better than conventional learning for class VIII students of MTsN 3 Lima Puluh Kota—the type of pre-experimental research with a Static Group Comparison Design research design. The population of the study was all students of class VIII MTsN 3 Lima Puluh Kota, consisting of 3 classes. The research sample consisted of students from class VIII.3, as an experimental class, and students from class VIII.2, as a control class, using the simple random sampling technique. The research instrument tested students' mathematics learning outcomes through essay questions—data analysis techniques with percentages for student learning activities and t-tests for students' mathematics learning outcomes. Based on the results of the analysis and student learning activities, an average rate of 90.56% was obtained with the criteria of very active. Meanwhile, from the analysis of students' mathematics learning outcomes data, it was obtained that tcount > ttable, namely tcount = 11.50 and ttable = 1.67. The testing criteria are accepted H0 if tcount > ttable; otherwise, reject H0. Based on the calculation, it was obtained that tcount > ttable, namely 11.50 > 1.67, meaning that H0 was rejected and H1 was accepted, so the Auditory intellectual repetition (A.I.R.) learning model is better than conventional learning in class VIII MTsN 3 Lima Puluh Kota.