Hi, this is author Kbr, I am back and I bring you new version of KoreanDollLikeness. (Please double check the naming of the LORA file when using sample images generation data, Civitai love to rename the LORA file automatically into something different) If you are wondering where are the older versions of my other LORAs( KDL, JPL, TDL), you can find them here on huggingface: https://huggingface.co/Kanbara/doll-likeness-series My Twitter: https://twitter.com/KbrLoras My Pixiv: https://www.pixiv.net/en/users/92373922 I usually post my works on Twitter and Pixiv. You can now use this model at: - SinkIn - DazzleAI (Link to discord: https://discord.gg/pq2RmzfaTx) - Tensor.art (free online generation) Please consider supporting me at: https://ko-fi.com/kbrloras It takes time and effort to collect training materials, training, testing LORAS, I would appreciate your support! ---------------------------------------------------------------------------------- Introduction 1. This version or KDL does not necessarily mean it is better or worse than v10 or v15, it is a different version that feature another set variety of faces. Recommended settings: Make sure you are aware on the usage instructions of LORA VAE: vae-ft-mse-840000-ema-pruned (For realistic models) Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers) Steps: 20 to 35 Clipskip: 1 or 2 Upscaler : Latent (bicubic antialiased) CFG Scale : 5 to 9 LORA weight for txt2img: anywhere between 0.2 to 0.7 are recommended Denoise strength for img2img: 0.4 to 0.7 It is also recommended to mix this LORA with my other DollLikeness LORAs for interesting effects ---------------------------------------------------------------------------------- License This model is open access and available to all, with a CreativeML OpenRAIL-M license further specifying rights and usage. The CreativeML OpenRAIL License specifies: You can't use the model to deliberately produce nor share illegal or harmful outputs or content The authors claims no rights on the outputs you generate, you are free to use them and are accountable for their use which must not go against the provisions set in the license You may re-distribute the weights and use the model commercially and/or as a service. If you do, please be aware you have to include the same use restrictions as the ones in the license and share a copy of the CreativeML OpenRAIL-M to all your users (please read the license entirely and carefully) (Full license here: https://huggingface.co/spaces/CompVis/stable-diffusion-license) Additional Add-ons to license/notes You shall take full responsibility for any creative work that uses this model Refrain from using this model for malicious intent, harm, defamation, scam or political usages. It may impair and discourage the author from producing more works. Disclaimer Creation of SFW and NSFW images is user's decision, user has complete control over to generate NSFW content whether or not. ---------------------------------------------------------------------------------- FAQ Q: Why can't I produce the same pictures as you? A: Sorry I don't share my prompt, you may check the recommended settings, but you may ask me for advice on Twitter or Pixiv Q: Where did your first Civitai account went? why did it disappear? A: I decided to delete the account due to news of controversial scam reported by online media outlet, I hope it will not happen again. Q: What is the difference of each version upgrade? A: Version upgrade does not mean it will fix hands or legs, it is mainly difference of the face of the LORA, newer versions have wider range of faces. Q: Will you release all your other LORAs? A: Yes, maybe, but I would like to keep certain LORA to be exclusive to fans or supporters in the future through fanbox/ko-fi Q: Do you take requests or commissions on making custom LORAs? A: I might do community voting for requests, if you are somehow interested in a custom/exclusive LORA, you may contact my through Pixiv or Twitter for discussion