v4_noobAIXL_v-pred-075sMerged based on noobAIXL_v-pred-075s.Recommended: masterpiece, best quality, newest, anime screenshot, absurdres, highresNegative: worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form feral, semi-anthroCFG: 3-7<hr />RollbackThis model is one of the ones I merged based on NoobXL_0.6 in my v-pred model testing.I am not satisfied with my recent merge model, so I will roll back to a merge model from a while ago.I personally found this model very interesting.You can start with no negative prompts.Use “worst quality” if you need it, and if the colors look funny, try adding “light” or something similar to the negative.You might want to put “masterpiece” or “anime screencap” in the positive prompt.”absurdres, highres, ai-generated, ai-assisted” is worth trying as a quality tag.I have no idea how I made this model myself.CFG: 2~<hr />v3.1_v-pred_065sTrained illustrious_0.1 with 10000 animated images, batch size 15, 12000 steps, and merged it into susamix_v2.Recommended prompt: anime screencapNegative prompts: worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthro(It might be more interesting to use only “worst quality”)CFG: 3-5<hr />v2_v-pred_065sAnimated model merged based on NoobAI-XL V-Pred-0.65SRecommended prompt: anime screencapNegative prompts: worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthroCFG: 1-7When using A1111, turning on the following extension may improve colors (used in the sample image)<a target="_blank" rel="ugc" href="https://github.com/ljleb/sd-webui-freeu.git">https://github.com/ljleb/sd-webui-freeu.git</a><hr />V-pred test 2Adjusted the model before merging and enhanced the animation painting.No need to include “anime screeccap” or “anime coloring” in the prompts.<hr />V-pred testThis model is V-Pred model.Simple animated model based on NoobAI-XL_V-Pred-06.Very simple.Recommended prompts: masterpiece, best quality, newest, absurdres, highres, anime coloringNegative prompts: worst quality, old, early, low quality, lowres, signature, username, logo, bad hands, mutated hands, mammal, anthro, furry, ambiguous form, feral, semi-anthroCFG: 4-5<hr />v1.0The “FreeU” extension was enabled when the sample images were generated.サンプル画像の生成時は「FreeU」拡張機能を有効にしていました。<hr />v1.0It is an animated model merging various models based on NoobAIXL.Recommended prompts: anime screencap, masterpieceNegative prompts: worst quality, bad qualityCFG scale: 3NoobAIXLをベースにいろいろなモデルをマージしたアニメモデルです。推奨プロンプト：anime screencap, masterpieceネガティブプロンプト：worst quality, bad qualityCFG scale: 3

susamix4_NoobAI

susamix4Noobai_v10.safetensors

v1.0

v4_noobAIXL_v-pred-075s

Rollback_noobAI-v-pred0.6

v3.1_v-pred_065s

v2_v-pred_065s

V-pred_test_2

V-pred_test

GX-mix_IL_动漫

该模型偏可爱风，以光滑动漫和动漫截图风格为特点；
使用“动画上色：Anime coloring”可以获得更详细和丰富的色彩。
正面tags:  
very awa, masterpiece, best quality, highres, absurdres, newest, aesthetic,
推荐设置：
VAE：已融入，无需添加
CFG Scale: 4 to 6
Sampler: Euler a  采样器：Euler a
Sampling steps: 25 to 30 采样步骤：25至30
Upscale by: x1.5 to 2.0 (or custom 768x1152->1024x1536) 会员：x1.5至2.0（或自定义768x1152-％3E1024x1536）
Upscaler: ESRGAN_4x/R-ESRGAN 4x+ Anime6B  放大器：ESRGAN_4X/R-ESRGAN 4x+ Anime6B
Denoising strength: 0.4  降噪强度：0.4
Hires steps: 15  步骤：15
ADetailer: 随意

IL-GX-mix4_v10.safetensors

Files with SHA256: 7ac153265427ab3579f52a8cf6365967762f14edd351ff3191356607d68ccfd0