Abstract. This study aims to determine the magnitude of the effect of the integration of direct learning models and interactive multimedia-based number beads on the learning outcomes of the fourth-grade students at the Oekabiti Catholic Elementary School, Kupang Regency. This study uses a quasi-experimental method with a posttest-only design type with the non-equivalent control group. There are 20 students taken as the population of this study, while the sample collection technique is achieved by using the census method, so the number of samples is 20 students. The instrument applies study documentation and tests. However, before the test instrument is used, the validity, reliability, distinguish, and difficulty index of the questions will be tested to obtain the validity of the instrument. The data analysis technique uses the independent t-test, but before the t-test is carried out, the data analysis requirements are tested, namely normality and homogeneity tests. The results showed that the test instrument in the form of 20 test items was proved to be valid, reliable. It also had good discriminating power and an intermediate level of difficulty. Meanwhile, the results of the normality and homogeneity tests showed that all data were normally and homogeneously distributed. The results of the hypothesis test showed that there is a difference in the average value between the control class and the experimental class which indicates that the experimental class is better at using a direct learning model assisted by interactive multimedia-based interactive media than that of the control class. The mean posttest value of the experimental class was 86.25 while the control class was 77.00, meaning 82.25 > 77.00, and based on the independent test results, the sample t-test showed a significance value of 0.000 <0.005. Thus, it can be concluded that there is a significant influence of the integration of the direct learning model with interactive multimedia-based number beads on the learning outcomes of the fourth-grade students. Keywords: of direct learning models, number beads, interactive multimedia Abstrak. Tujuan dari penelitian untuk mengetahui besarnya pengaruh integrasi model pembelajaran langsung dan manik – manik bilangan berbasis multimedia interaktif terhadap hasil belajar siswa kelas IV SD Katolik Oekabiti Kabupaten Kupang. Penelitian menggunakan metode kuasi eksperimen dengan jenis desain posttest only with non equivalent control group. Populasi berjumlah 20 siswa, sedangkan teknik pengambilan sampel menggunakan metode sensus, sehingga jumlah sampel sebanyak 20 siswa. Instrumen menggunakan studi dokumentasi dan tes. Namun sebelum instrument tes digunakan akan diuji validitas, reliabilitas, daya pembeda, dan tingkat kesukaran soal untuk mendapatkan keabsahan instrument. Teknik analisis data menggunakan uji independent t-test, namun sebelum dilakukan uji t dilakukan uji persyaratan analisis data yaitu uji normalitas, homogenitas. Hasil penelitian menunjukkan bahwa instrumen tes sebanyak 20 soal dinyatakan, valid, reliabel, memiliki daya pembeda soal yang baik serta memiliki tingkat kesukaran soal yang sedang. Sementara itu hasil uji normalitas dan homogenitas menunjukkan semua data berdistribusi normal dan homogen. Hasil uji hipotesis menunjukkan ada perbedaan nilai rata-rata antara kelas kontrol dan kelas eksperimen yang menunjukkan bahwa pada kelas eksperimen lebih baik menggunakan model pembelajaran langsung berbantu media manik-manik bilangan berbasis multimedia media interaktif daripada pada kelas kontrol. Rata – rata mean nilai posttest kelas ekperimen 86,25 sedangkan kelas kontrol sebesar 77,00, artinya 82,25 > 77,00 serta berdasarkan hasil uji independet sample t-test menunjukkan nilai signifikansi 0,000 <0,005. Dengan demikian, dapat disimpulkan bahwa terdapat pengaruh integrasi model pembelajaran langsung yang signifikan dengan manik – manik bilangan berbasis multimedia interaktif terhadap hasil belajar siswa kelas IV. Kata Kunci: Integrasi model pembelajaran, manik-manik bilangan, multimedia interaktif