Multi-modality image fusion is a technique that integrates complementary information from multiple imaging modalities, particularly enhancing functional highlights and texture details. A critical challenge in this domain is how to effectively extract and fuse the features from different modalities while fully exploiting their complementary information. To tackle this issue, we propose the Cross-Modal Multi-scale Attention for Infrared and Visible Image Fusion (CMA) for image fusion. we propose a novel method called Cross-Modal Multi-scale Attention (CMA) for Infrared and Visible Image Fusion. The proposed CMA framework introduces an innovative attention mechanism along with a multi-scale feature fusion module. This approach enables the effective extraction and integration of features from different modalities, thereby enhancing fusion performance. These components are designed to capture information at different scales while improving the richness and accuracy of the feature representation. Extensive experimental results demonstrate that CMA outperforms existing methods in multiple fusion tasks such as infrared-visible image fusion and medical images.