In this paper, two present compact models for generic undoped double-gate (DG) MOSFETs are compared and discussed from aspects of formulations, advantages, prediction result, and computational efficiency. The models under comparison make fundamental contributions to generic DG MOSFET modeling. However, the applications of these models are hindered by their own drawbacks which are discussed in this work. The comparison results will provide a very useful reference and may benefit the future work for researchers to improve the development of double-gate MOSFET compact model. In addition, the calculation efficiency is extensively improved based on the Jacobian-Newton iterative method with a universal initial guess. (C) 2010 Elsevier Ltd. All rights reserved.