<ver.1.0>

Features of this workflow are as follows:

Images are regenerated using Z Image Turbo.

Image captions are automatically generated using Florence, but if you would like to use your own text, change the switch to "true."

If you are using ControlNet instead of i2i, change i2i / ControlNet Switching to "true." Also, activate only the Canny, OpenPose, and Depth nodes you want to apply (bypass the others), and connect them to the "Image" node of the "QwenImageDiffsynthControlnet" node.

Simply drop the json file or attached image into ComfyUI to use it.

Have a great image generation life!