Single-image deblurring is a key foundation in low-level visual enhancement tasks. However, existing research has overlooked the role of image depth and frequency information in enhancing deblurring and model generalization. In this paper, we explore and verify the synergistic effect of depth information and frequency information on single-image deblurring. Firstly, we used the Depth-Anything Model to obtain the depth map corresponding to the blurred image. Then, we cascade a lightweight depth-aware module to fuse the blurred image input with the depth map and a frequency module to enhance frequency information. The two modules proposed in this paper significantly improve the deblurring ability of the state-of-the-art models, with the PSNR value increased by 0.15–0.35. The proposed modules also improve the model’s generalization ability across different datasets.