This paper introduces new model parameterizations for learning dynamical systems from data via the Koopman operator, and studies their properties. Whereas most existing works on Koopman learning do not take into account the stability or stabilizability of the model -- two fundamental pieces of prior knowledge about a given system to be identified -- in this paper, we propose new classes of Koopman models that have built-in guarantees of these properties. These models are guaranteed to be stable or stabilizable via a novel direct parameterization approach that leads to unconstrained optimization problems with respect to their parameter sets. To explore the representational flexibility of these model sets, we establish novel theoretical connections between the stability of discrete-time Koopman embedding and contraction-based forms of nonlinear stability and stabilizability. The proposed approach is illustrated in applications to stable nonlinear system identification and imitation learning via stabilizable models. Simulation results empirically show that the learning approaches based on the proposed models outperform prior methods lacking stability guarantees.