I trained the model with a small set of images featuring a single character, allowing it to function as a character model (where only the character's name is input, without the "3D style" tag) as well as a 3D style rendering model (where only the tag named "3D style" is input, without the character's name). It can be used in combination with other characters lora
