Optimized and improved based on the Text-to-Image foundational workflow, supports Chinese input prompts and ControlNet, with clear processes, easier to use, and more practical