VersaCamVLA consistently outperforms prior VLA methods and direct multi-view baselines on multiple tasks, and maintains stable performance under changes in the number of cameras and unseen camera poses.