v1.0 Lora has been trained with the ntr4.0 model.These images are the result of output from comfyui to the waiNSFWIlustrious_v70 model.Output is also possible with other Illustrate-xl-based models.Character prompt : bishopfull prompt :positive : 1girl,bishop,blonde hair,long hair,hair between eyes,large breasts,cleavage,cleavage cutout, clothing cutout, diadem,dress, hood,hooded robe, open clothes, open robe,robe,white dress,white hood, white robeme,dium breasts, open mouth,smile,completely nude,masterpiece, best quality, newest, absurdres, highresnagative : lowres,bad anatomy,blurry,(worst quality:1.8),low quality,hands bad,(normalquality:1.3),bad hands,mutated hands and fingers,extra legs,extraarms,duplicate,cropped,jpeg,artifacts,long body,multiple breasts,mutated,disfigured,badproportions,bad feet,ugly,text font ui,missing limb,monochrome,face bad,resolution : 832 x 1216step : 35cfg : 5.0sampler : euler11/23 update xl_loraI made another one for xl-Illustrious general purpose. Use the same prompt as before11/24 update xl_lora (ver 2)I've adjusted the color. It's more of an original hair color, and the test has eliminated the problem of not getting satisfactory results like in-game 3D modeling if it's not the waiNSFWIlustrious_v70 model. The ilustratrious_xl still doesn't do well, but the derivatives do. The bad thing about this Lora is that it comes out a little blurry.I recommend that you change the prompt from blonde hair to light blonde hair.12/02 update NTR 4.0(ver2)It was made in a mixed form of ILXL ver2 and NTR ver1.I was satisfied when I printed out the adapter with a value of 0.2, so I don't think there will be any more uploads.12-29 pmxI trained by adding the data extracted from the game model file. I uploaded it for personal backup purposes.
