This chapter examines the capability of Support Vector Machine (SVM) for prediction of ultimate capacity (Q) of driven piles in cohesionless soils. SVM that is firmly based on the theory of statistical learning theory, uses regression technique by introducing ε-insensitive loss function has been adopted. SVM achieves good generalization ability by adopting a structural risk minimization (SRM) induction principle that aims at minimizing a bound on the generalization error of a model rather than the minimizing the error on the training data only. SVM is trained with optimization of a convex, quadratic cost function, which guarantees the uniqueness of the SVM solution. In this chapter, the developed SVM model outperforms the artificial neural network (ANN) model based on root-mean-square-error (RMSE) and mean-absolute-error (MAE) performance criteria. An equation has been developed for the prediction of Q of driven piles based on the developed SVM model. A sensitivity analysis has been also done to determine the effect of each input parameter on Q. This chapter shows that the developed SVM model is a robust model for prediction of Q of driven piles in cohesionless soils. © 2011 by Nova Science Publishers, Inc. All rights reserved.