SDXL
"7th_anime_XL-IL" Series Model Description & RecipeModel Overview & PhilosophyThis series is fundamentally an attempt to improve upon existing anime models through a "brute force" methodology.The creation method involved tens of thousands of trials to select the best possible combinations, focusing on models that demonstrated the highest responsiveness to prompts and the fewest anomalies in generating hands and fingers.Model Characteristics7th_anime_XL-IL_A: Features a flatter coloring style.7th_anime_XL-IL_B: Features a rich/deep coloring style.7th_anime_XL-IL_B2: Features an even richer/deeper coloring style than B.7th_anime_XL-IL_C: Features a more realistic coloring style, which can sometimes produce good results for sensitive content.Creation Recipe (Simplified Version)1. Creation of Common Component "D4343A"Base Model: Illustrious-XLFinetuning: Finetuned for 1 epoch using a dataset of 151,345 images.2. 7th_anime_XL-IL_A Creation RecipePerform a block weighted merge of illustriousXLPersonalMerge_v30Noob10based and illustrijEVO_lvl3 with the following settings:0,0,0.0740740740740741,0.259259259259259,0.5,0.740740740740741,0.925925925925926,1,0.925925925925926,0.740740740740741,0,0.0740740740740741,0.259259259259259,0.5,0.740740740740741,0.925925925925926,1,0.925925925925926,0.740740740740741,0.57th_anime_XL-IL_A is created by performing a block weighted merge of waiNSFWIllustrious_v140 and the model created in Step 1, using the following settings:0,1,0.984809027777778,0.940972222222222,0.87109375,0.777777777777778,0.663628472222222,0.53125,0.383246527777778,0.222222222222222,0.5,0.687065972222222,0.871527777777778,0.94921875,0.777777777777778,0.616753472222222,0.46875,0.336371527777777,0.222222222222222,0.128906253. 7th_anime_XL-IL_B Creation RecipePerform a block weighted merge of hassakuXLIllustrious_v30WIP and 7th_anime_XL-IL_A with the following settings:0,0,0.0740740740740741,0.259259259259259,0.5,0.740740740740741,0.925925925925926,1,0.925925925925926,0.740740740740741,0,0.0740740740740741,0.259259259259259,0.5,0.740740740740741,0.925925925925926,1,0.925925925925926,0.740740740741,0.5Extract the difference (delta) between D4343A (from "1. Common Component") and its base model Illustrious-XL to create a LoRA.7th_anime_XL-IL_B is created by applying the LoRA from Step 2 to the model from Step 1 with a strength of 0.6.4. 7th_anime_XL-IL_B2 Creation RecipePerform a weighted merge of 7th_anime_XL-IL_B and ntrMIXIllustriousXL_xiii at a 0.6 : 0.4 ratio.7th_anime_XL-IL_B2 is created by performing a block weighted merge of the model from Step 1 and ntrMIXIllustriousXL_xiii with the following settings:0,0,0,0,0,0,1,1,1,1,0,0,0,1,1,1,1,1,0,05. 7th_anime_XL-IL_C Creation RecipePerform a block weighted merge of illustriousXLPersonalMerge_v30Noob10based and illustrijEVO with the following settings:0,0,0,0,0,0,0,1,1,1,0,0,1,1,1,1,1,0,0,07th_anime_XL-IL_C is created by performing a block weighted merge of hassakuXLIllustrious_v30WIP and the model from Step 1 with the following settings:0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1All Models using the Fair AI Public License 1.0-SD license.
