<p>Utilizing recycled concrete aggregates in concrete is a sustainable method to reduce the need for extracting natural mineral resources and lessen the negative environmental impacts of the concrete. However, it faces hurdles since the hardened mortar sticking to natural aggregates is vulnerable. This results in a higher vulnerability to cracking and decreased strength. Hence, the present article applied the least square support vector regression (LSSVR) framework is taken into account in order to achieve this. The Artificial rabbit algorithm (ARA) and Golden jackal algorithm (GJA) are integrated with the LSSVR framework to enable hyperparameter tweaking and the identification of the best-performing combination of hyperparameters. Drawing from peer-reviewed published research, a collection of data involving 257 numbers and 10 inputs (Water, cement, natural concrete aggregate content, recycled concrete aggregate content, superplasticizer, maximum aggregate size of RCA, the density of RCA, water absorption of RCA, fibers, and fiber type (Steel fiber, Carbon fiber, Polypropylene fiber, Basalt fiber, Glass fiber, and Woolen fiber) was randomly divided into 3 stages: testing (15%), validating (15%), and training (70%). The findings indicated that the LSSVR_ARA and LSSVR_GJA approaches have excellent promise for accurately predicting the STS of fiber reinforced RAC. Regarding the values of the measured metrics and justifications, it was observed that LSSVR_GJA depicted the highest accuracy and acceptable reliability and uncertainty, causes recognizing superior model.</p>