A driving model training method, a driver recognition method and apparatus, a device, and a medium. The driving model training method comprises: obtaining training behavior data of a user, the training behavior data being associated with a user identifier (S11); obtaining, according to the training behavior data, training driving data associated with the user identifier (S12); obtaining, according to the user identifier, positive and negative samples from the training driving data (S13); and training an error backpropagation neural network model by using the positive and negative samples to obtain a target driving model (S14). The driving model training method resolves the problem of poor recognition effect of an existing driving model, and improves the accuracy of driver recognition