Eventually consistent is a concept in distributed computing that describes a model where data changes are propagated to all nodes in a system over time. Unlike strong consistency, where all nodes reflect the same data immediately, eventually consistent systems allow for temporary discrepancies. This means that while different nodes may have different versions of data at any given moment, they will converge to the same value eventually. This approach is often used in systems like NoSQL databases and cloud storage, where high availability and partition tolerance are prioritized. It allows for better performance and scalability, as updates can be made without waiting for all nodes to synchronize immediately.