Inférence
In brief
The moment when a pre-trained AI model sets to work to respond to your request. Training is the learning process, once and for all. Inference is the application of that learning, with every request.
▶ Precise definition
The phase in which a trained model is used to generate an output from an input. Distinct from training, it represents an ongoing cost and raises issues of sovereignty when it is to be carried out within a given territory, under a given legal framework, on equipment under that jurisdiction’s control.
See also