In view of the problem of low computational efficiency in multi-mode Synthetic Aperture Radar(SAR) imaging processing, a parallel acceleration method is proposed for multi-mode SAR imaging based on Graphic Processing Unit(GPU). In order to make full use of GPU's memory resources and improve the efficiency of the algorithm, in the parallel computing part of the algorithm, the large-scale data parallel is carried out in the matrix transposition and matrix multiplication by using shared memory. The experimental results show that the algorithm greatly improves the efficiency of multi-mode SAR imaging, and the maximum acceleration ratio reaches 55.62, which solves the problem of low utilization of GPU.