We develop two models for Myxobacteria swarming, a modified Lattice Gas Cellular Automata (LGCA) model and an off-lattice CA model. In the LGCA model each cell is represented by one node for the center of mass and an extended rod-shaped cell profile. Cells check the surrounding area and choose in which direction to move based on the local interactions. Using this model, we obtained a density vs. expansion rate curve with the shape similar to the experimental curve for the wild type Myxobacteria. In the off-lattice model, each cell is represented by a string of nodes. Cells can bend and move freely in the two-dimensional space. We use a phenomenological algorithm to determine the moving direction of cells guided by slime trail; the model allows for cell bending and alignment during collisions. In the swarming simulations for A+SMyxobacteria, we demonstrate the formation of peninsula structures, in agreement with experiments.