The training images are from the public network, this model is only for learning, do not use for profit and illegal ways.
This model is beta version, image generation if there are errors and problems, please try to add more and more accurate negative descriptions, otherwise it is not accurate.
This model is not stable, please reduce the weight as much as possible, otherwise the picture will be blurred, sometimes there will be bodies or bloody, scary faces and other uncomfortable pictures, please use carefully.
It is recommended to use a red theme because other colors may not be accurate, and try not to use too many tags, which may not be recognized by the model and it is difficult to produce good results.