Ultrafine graded fly ash (UFGFA) is a well-graded pozzolanic material among silica-rich reactive pozzolans that contributes to the improvement of the early pozzolanic reactivity of cement paste, mortar, and concrete at early ages. The purpose of this article is to demonstrate the effect of UFGFA on the development of compressive strength in M20, M25, and M30 grade concretes with a partial replacement of cement ranging from 0% to 40% by weight at various water-binder (W/B) ratios of 0.43-0.53. Compressive strength development was examined at ages 7, 14, 28, and 56 days. Additionally, three machine learning (ML) models (linear regression, ANN, and GEP) were used to predict the compressive strength of the concretes using different predictive classifications.According to the investigations, experimental studies indicate that a lower W/B ratio and a minimum amount of UFGFA replacement result in an increase in strength at an early age of 7 days. At 14, 28, and 56 days, higher strength results were obtained when 15%-30% UFGFA was replaced with a W/B ratio less than 0.5, whereas a reduction in strength was observed when more than 30% UFGFA was replaced with cement and with W/B ratio greater than 0.5. All machine learning models produced good predictions of compressive strength, with LR models having least performance and ANN has the more accurate fit. ANN models outperformed the other three machine learning models, but GEP models performed identically near to the ANN model, with expressions generated, that can be used for further development of predictive model.artificial neural net, fly ash, genetic algorithm, linear regression, pozzolan, ultrafine graded fly ash Discussion on this paper must be submitted within two months of the print publication. The discussion will then be published in print, along with the authors' closure, if any, approximately nine months after the print publication.