The connectivity model provides a crucial decision-making basis for conservation planning, including nature reserves, ecological networks, and ecological security patterns. Currently, the mainstream connectivity model is to construct network systems composed of ecological patches and corridors based on binary landscapes. However, during the modeling process, factors such as uncertainties in characterizing species movement, subjectivity in defining multiple thresholds, and homogeneity in simplifying model elements limit the applicability of this kind model in meeting the needs of ecological conservation planning, including multi-scale integration, multi-species conservation, and adapting to climate change. To address these limitations, in academia, some research has begun to shift towards exploring the gradient connectivity model oriented towards whole landscape-wide domains, namely omnidirectional connectivity. This study has systematically reviewed the progress of research in terms of theoretical foundations, implications, algorithms, and applications of omnidirectional connectivity model, and conducted a comprehensive comparison with current mainstream network-based connectivity model. This study finds that, as a relatively novel connectivity modeling framework, the omnidirectional connectivity model has been widely applied in the field of biodiversity conservation and applied to measure the progress of other landscape diffusion, such as the spread of wildfires and the penetration of urban green spaces. It also exhibits significant advantages in terms of multi-scale integration, spatial precision, and result validation. Finally, based on the characteristics and advantages of the omnidirectional connectivity model, potential future research directions and recommendations are proposed in terms of model uncertainty, model coupling and integration, and species movement observation.