I use the Taniguchi san style model with other style Lora models to generate images, and then filter the colored images with Taniguchi style as a new training set to make up for the defect that the original model can only draw grayscale images. This method is more time-consuming and laborious, but the effect is very good,In addition, I am still adjusting the compatibility of the skinsuit model with other models. It has a great impact on the style of paintingI will keep updating it.