An end-to-end guide and investigation into Stable Diffusion tooling — SD WebUI, models, LoRAs, and VAEs — for generating synthetic training data for military computer vision models.