Inference:

(as used in MLOps; aka prediction) The process of sending new data to a trained (and deployed) model and getting predictions back. In academic literature inference often has a different meaning.