Air quality is a comprehensive indicator that reflects air pollution. The air quality index is defined as a description or value of the transformation of individual parameters of interrelated air pollution, such as PM10, SO2, CO, O3, NO2 into one value or a set of values so that it is easy to understand for the general public. Therefore, predictions of the Air Pollutant Standard Index in the future are needed for future policy decision making. SVR is a development of Support Vector Machine (SVM) for regression cases. The aim of this study is to predict the Air Pollutant Standard Index (ISPU) in the future using the SVR method. In the SVR method, the best kernel is used as an aid in solving non-linear problems, the Min-Max Normalization method for data normalization, division of training and testing data, selection of the best model with Grid Search Optimization. The best prediction results were obtained using a radial kernel with values with parameters ε = 0.1, C = 10, and γ = 1 with the smallest error of 0.0086, with an RMSE of 0.0894. The RMSE value indicates that the model's ability to follow data patterns well. From the model on the radial kernel, the predicted results of the Air Pollution Standard Index from January 1, 2025 to July 31, 2025 were obtained which were not constant or fluctuating with an interval range of 41,14 – 58,69 and based on the Air Pollution Standard Index category, it was in the fairly good category index.