A LoRA that has been trained to generate fairly accurate images of the character Karyl from the anime and mobile game Princess Connect! Re:Dive. Like my previous LoRA, it is not trained for NSFW but performs relatively well when used with NSFW prompts. It was created based on the NovelAI model, but also works well with the AnythingV3 model, as featured in a few of the images. Other models have not been tested, but shouldn't cause much of an issue. The dataset provided contained 30 images of the character featured in official art, anime scenes, with a few works from unaffiliated artists.
Note: Sometimes it likes to try to produce images of Karyl with slightly pinkish hair and eyes that lean toward a yellowish-green. This can be fixed by adding the prompts "black hair" and "green eyes". It also doesn't always add cat ears 100% of the time, so you can use the prompt "cat ears" to specify them.
Good prompts to use:
Masterpiece, best quality, highres, cat ears, black hair, green eyes,
Negative prompts to use:
lowres, bad anatomy, bad hands, text, error, missing fingers, extra digit, fewer digits, cropped, worst quality, low quality, normal quality, jpeg artifacts, signature, watermark, username, blurry, bad face, ugly face, extra limbs, bad arms, ugly eyes, extra tail
描述:
训练词语:
名称: Karyl.safetensors
大小 (KB): 147572
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
