This paper presents a hybrid distributed data mining (DDM) model for optimization of response time. The model combines a mobile agent approach with client server strategies to reduce the overall response time. The hybrid model proposes and develops accurate a priori estimates of the computation and communication components of response time as the costing strategy to support optimization. Experimental evaluation of the hybrid model is presented.