According to previous research, porous concrete has lesser compressive strength than normal concrete and can only withstand mild traffic loads. The demand for high-strength porous concrete is urgent. By balancing the elements of porous concrete, high strength porous concrete may be produced. This may be accomplished by comprehending the complicated interaction between porous concrete's constituents. Using artificial intelligence approaches, the present research offers a study on the prediction of 7, 28-day compressive strengths and 28, 90- day flexural strengths of porous concrete. Random tree (RT), M5 Rules, REP Tree, and ensemble with Bagging were used to estimate the compressive strength of porous concrete in this work, and a comparison of performance and practicality of artificial intelligence result is shown. A total of 111 datasets were extracted from prior research and publications. As input parameters for modelling, cement, natural coarse aggregates, recycled coarse aggregates, water, silica fume, plastic/steel/rubber fibres, and density were employed. Properties such as mean absolute error, root mean square error, and correlation coefficient were used to evaluate and compare the behavior of projected models. The results demonstrated a high level of capacity and potential for predicting the compressive of porous concrete.