 This paper proposes a novel optimization methodology for antenna design. The proposed methodology uses a variable resolution EM simulation model to optimize antennas. The model shares the same physical background and is selected from a continuous spectrum of allowable resolutions. The lower resolution model is used during the initial stages of the search process, while the high resolution model is used later on. The optimization algorithm is based on a particle swarm optimizer and is validated using various antenna structures of different characteristics. The results show that the proposed methodology can achieve substantial computational savings compared to high fidelity based optimization while maintaining similar reliability. Furthermore, the proposed methodology is easy to implement and versatile. This article was authored by Slava Mikozil and Anna Piatrenko-Dabreska.