In order to improve the precision of selective laser sintering(SLS) and to solve the high cost of technical parameters optimization,five technical parameters of laser power,preheating temperature,scanning speed,scanning distance and layer thickness were selected to design the orthogonal experiment to obtain sample data and a unified target function was established.By using the seeker optimization algorithm(SOA),the least squares support vector machine(LSSVM) was optimized and the accurate predictive model of SLS based on SOA-LSSVM was established;the unified performance of the parts under different technical parameters combinations was predicted,and compared with the results obtained by using traditional BP neural network and LSSVM model.The results show that the SOA-LSSVM model has good generalization ability for small sample prediction and the maximum relative error between the predicted value and the actual value is only 1.11%,which can provide reference for the selection of SLS processing parameter combinations.