Analog filters have been successfully used in many real-life applications Minimizing group delay ripple can be very challenging so metaheuristic algorithms are often needed to deal with this problem. In this paper, four different metaheuristic algorithms, namely backtracking search, grey wolf optimizer, lightning search and multi-verse optimizer algorithms, have been applied for optimization of the analog filter group delay. The evaluations of the algorithms performances are tested using the fifth-order Chebyshev low-pass filter as a base filter, then the second, third and fourth order all-pass filter structures are connected to it in cascade form respectively. Afterwards, group delay of the filter is minimized for the each cascaded all-pass filter structure. The group delay responses and the optimal parameters of the optimized filters have been determined. Comparisons have been made with the metaheuristic algorithms as well as other design techniques published in the literature to analyze which method works more efficiently on which filter structure.