Fly ash is produced as a byproduct of the coal combustion process in thermal power plants. Fly ash consists of very fine and microscopic particles, typically composed of mineral compounds such as silicon dioxide, aluminum oxide, and iron oxide. These compounds make fly ash suitable for use in various industries, particularly in the construction industry. Applications of fly ash include additives in concrete, fillers in asphalt, production of bricks and concrete blocks, and pollutant absorption. As a pozzolanic material, fly ash helps reduce carbon dioxide emissions in the cement production process. In this study, a comprehensive database of previous studies on fly ash concrete was initially collected. This data included 599 samples from credible laboratory studies. The gathered dataset consisted of various input variables, including the water-to-cement ratio, amount of fly ash, cement content, coarse aggregate amount, fine aggregate amount, superplasticizer content, and curing age of the concrete. To predict the compressive strength of the concrete, various machine learning algorithms were utilized, including Genetic Programming (GP), Adaptive Neuro-Fuzzy Inference System (ANFIS), Multi-Layer Perceptron (MLP), Radial Basis Function Neural Network (RBF), Kriging, and Extreme Learning Machine (ELM). Furthermore, the accuracy of each model was evaluated using statistical indices, and the best model was identified. The results show that different machine learning models exhibit varying performances in predicting compressive strength. In particular, the Kriging method, with a correlation coefficient of 0.96, was selected as the best model.