1、This model was trained on screenshots taken from within the game, It exhibits high levels of similarity and generalization.
2、To enhance the expressiveness and liveliness of the generated characters, it is recommended to use emotional and action descriptors in the prompt. Furthermore, descriptions from different perspectives might produce surprising results.
3、If the desired results are not achieved, try increasing or decreasing the weight of any prompt that might be causing issues.
A reminder:
this Lora model is not suitable for beginners. Simple prompts often result in less than ideal outcomes. It is recommended to use it in conjunction with inpainting and complex, meticulous prompts.