* [Push vs Pull Model](getting-started/architecture-and-components/push-vs-pull-model.md)

{% content-ref url="push-vs-pull-model.md" %}

[push-vs-pull-model.md](push-vs-pull-model.md)

{% endcontent-ref %}

Feast uses a [Push Model](https://en.wikipedia.org/wiki/Push_technology) (i.e., the feature store pushes feature values to the online store) to serve features in

real-time. This model is more efficient than a pull model (i.e., the model serving system pulls feature values from the

feature store) because it reduces the latency of feature retrieval.

In a [Pull Model](https://en.wikipedia.org/wiki/Pull_technology), the model serving system must make

a request to the feature store to retrieve feature values, which can introduce latency. In a push model, the feature

store pushes feature values to the online store, which reduces the latency of feature retrieval.