This paper proposes an efficient image deblurring network called ALGNet, which leverages a selective state space model (SSM) to aggregate enriched and accurate features. The key innovation is the introduction of an Aggregate Local and Global Information Block (ALGBlock), which combines a module for capturing local and global features (CLGF) and a feature aggregation module (FA). The CLGF module consists of two branches: a global branch that uses a selective SSM to capture long-range dependencies with linear complexity, and a local branch that employs simplified channel attention to model local connectivity, reducing local pixel forgetting and channel redundancy. The FA module dynamically recalibrates weights during feature aggregation to emphasize local information. Experimental results show that ALGNet outperforms state-of-the-art methods on widely used benchmarks, achieving superior performance with reduced computational costs. The method is efficient, scalable, and effective in restoring sharp images from blurred ones. The ALGNet model achieves state-of-the-art performance while maintaining computational efficiency, demonstrating its effectiveness in image deblurring tasks.This paper proposes an efficient image deblurring network called ALGNet, which leverages a selective state space model (SSM) to aggregate enriched and accurate features. The key innovation is the introduction of an Aggregate Local and Global Information Block (ALGBlock), which combines a module for capturing local and global features (CLGF) and a feature aggregation module (FA). The CLGF module consists of two branches: a global branch that uses a selective SSM to capture long-range dependencies with linear complexity, and a local branch that employs simplified channel attention to model local connectivity, reducing local pixel forgetting and channel redundancy. The FA module dynamically recalibrates weights during feature aggregation to emphasize local information. Experimental results show that ALGNet outperforms state-of-the-art methods on widely used benchmarks, achieving superior performance with reduced computational costs. The method is efficient, scalable, and effective in restoring sharp images from blurred ones. The ALGNet model achieves state-of-the-art performance while maintaining computational efficiency, demonstrating its effectiveness in image deblurring tasks.