This model has been exclusive to Tensor Art for a long time.
This version of the model was additionally trained for styles in 50,000 steps, which resulted in a significant improvement in understanding styles. Then, an additional combination of the DreamBooth-trained version with intermediate versions of the model was performed in order to soften the learning outcome and put the weights in their proper places. The model is capable of creating images in any style, including realistic ones. This is a joint project with the author of Triple_Headed_Monkey.
each image contains a WF and a prompt. Unfortunately Civitai cannot read these WF's