Precision medicine approaches have revolutionized oncology. Personalized treatments require not only identification of the driving molecular alterations, but also development of targeted therapies and diagnostic tests to identify the appropriate patient populations for clinical trials and subsequent therapeutic implementation. Preclinical in vitro and in vivo models are widely used to predict efficacy of newly developed treatments. Here we discuss whether, and to what extent, preclinical models including cell lines, organoids and tumorgrafts recapitulate key features of human tumors. The potential of preclinical models to anticipate treatment efficacy and clinical benefit is also presented, using examples in different tumor types.