This model is based on art therapy coloring images. Euler a sampling method produces the best results with this model. This LoRA model was developed and released solely for non-commercial research and educational purposes.

The model was trained on a diverse set of publicly available images collected from the internet to study fine-tuning. These images were not used for the purpose of reproduction, redistribution, or commercial use. While many of the images may originate from copyrighted sources, they were used under the assumption of fair use due to the transformative nature of the training and the non-commercial, academic scope of this work.

You should use Lineart to trigger the image generation.

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.