In this article, a hybrid method of adaptively modified characteristic basis function method (AMCBFM) and QR factorization with a dual modified Gram‐Schmidt (Dual‐MGS) algorithm, which is called QR‐AMCBFM, is presented. The first step is to derive the primary basis functions by using AMCBFM. To get the high‐level basis functions, QR factorization algorithm with Dual‐MGS is used to decompose the mutual coupling matrices. Different threshold values for the near and far field domains are set in the QR scheme. When the numerical rank gotten in the upper program is down to a set threshold, the rest interaction between farther blocks would be ignored. Subsequently, the high‐level characteristic basis functions and the coefficient of current are calculated. Once the adjacent currents precision arrives at a certain criterion, the higher‐level basis functions are thought of unnecessary. The hybrid algorithm estimates adaptively in the entire process. © 2007 Wiley Periodicals, Inc. Microwave Opt Technol Lett 49: 2879–2883, 2007; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mop.22830