Standard CLA-EC which is introduced recently is an evolutionary computing model obtained by combining cellular learning automata (CLA) model and evolutionary computing (EC) model. Some drawbacks of this model are low convergence speed and low accuracy for some optimization problems. In this paper a new version of CLA-EC called Continuous Action Set CLA-EC or in short Continuous CLA-EC is proposed. In this new version, the finite action set learning automaton in each cell is replaced by a continuous action set learning automaton. To show the effectiveness of the proposed model it is tested on some function optimization problems. The results of experimentations have shown that the proposed Continuous CLA-EC comparing to standard CLA-EC has both higher accuracy and speed of convergence.