In the computer experiments setting, Space-Filling Designs (SFDs) are widely used to explore the complex relationship between inputs and outputs. In this paper, a new SFD is initially defined with the help of the Strauss process. Through Markov chain Monte-Carlo (McMC) methods, more general Gibbs processes can be used to perform different goals. We will see that it is easy to sample over the entire range of each input variable as Latin hypercubes do it. Moreover, non-homogeneous designs can be constructed to take account of a priori information.