During the current pandemic, students are having difficulty understanding mathematical material; normally, they are taught by lecturers face-to-face, but now online. The obstacle that students face is that they believe that learning is not optimal because the signal is often poor and inconsistent, and they are unable to meet with the lecturer. To alleviate this issue, it is required to create a video of learning material that will be supplied by a mathematics lecturer by the lecturer who teaches, the video is published on YouTube, and the youtube link is given to students to study beforehand for the forthcoming meeting. This is consistent with the flipped classroom concept, in which students are expected to first understand the lecturer's video before being discussed directly by the lecturer.It is expected that students are able to think creatively about video content created by their professors, and students are able to examine material flaws that they have not understood and then clarified by their lecturers at the next meeting. In research on the development of instructional video media for high school mathematics curricular with the ADDIE stage model. This model, as the name suggests, consists of five main phases or stages, namely (A) analysis, (D) design, (D) development, (I) implementation, and (E) evaluation. The Implementation and Evaluation stage was carried out in the second year, with expanded tests carried out at UIN Walisongo Semarang in a fun and enjoyable manner, with posttests and user response questionnaires given with the following results: Based on research at UIN Walisongo, posttest results were processed using tests t results obtained are 0% <5%, then H0 is rejected. This means that we accept H1, which means that the results of learning mathematics using video learning media for high school mathematics curriculars are better than conventional learning models. Looking at the average learning achievement in the mean column, Group Statistics table, the experimental class average is 86.00, while the average -the average control class is 64.74. These results show that the learning outcomes of the experimental class are better than the control class. R Square value of 0.804 = 80.4%. This value means that the influence of virtual lab media on learning achievement is 80.4%, while 19.6% of learning achievement is influenced by other variables outside of the independent variables in this research.