Abstract: In this paper, a new non-monotone conic trust region method with fixed step-size based on conic models for solving unconstrained optimization problem is proposed. Unlike the traditional trust region methods, the sub-problem of our new algorithm is the conic minimization sub-problem. Moreover, we use the fixed step-size to obtain a new point when the trial step is not accepted. Theoretical analysis indicates that the new method preserves the global convergence under suitable conditions.