Type | |
Stats | 2,638 |
Reviews | (411) |
Published | Jul 28, 2023 |
Base Model | |
Usage Tips | Clip Skip: 2 |
Hash | AutoV2 C23CD0172C |
v1.4
有很多人提到之前的版本太黑了,画面太黑和发黄的问题都在这个版本被修复了,另外也改善了读取prompt的能力,比之前版本更听话了,同时丰富了一些背景元素,希望你能玩的开心~
Steps: 28, Sampler: DPM++ 2M SDE Karras, CFG scale: 9, Size: 512x768, Model hash: c23cd0172c, Model: POPE2.5_v1.49.fp16, Denoising strength: 0.4, Clip skip: 2, RNG: CPU, ADetailer model: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer dilate/erode: 32, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer version: 23.7.9, Hires upscale: 1.5, Hires steps: 12, Hires upscaler: R-ESRGAN 4x+ Anime6B, Eta: 0.667, Version: v1.4.1
There are a lot of people mentioned that the previous version is too dark, the screen is too dark and yellowing problems are fixed in this version, in addition to improving the ability to read the prompt, more obedient than the previous version, as well as enriched some of the background elements, I hope you will have fun~!
Steps: 28, Sampler: DPM++ 2M SDE Karras, CFG scale: 9, Size: 512x768, Model hash: c23cd0172c, Model: POPE2.5_v1.49.fp16, Denoising strength: 0.4, Clip skip: 2, RNG: CPU, ADetailer model: face_yolov8n.pt, ADetailer confidence: 0.3, ADetailer dilate/erode: 32, ADetailer mask blur: 4, ADetailer denoising strength: 0.4, ADetailer inpaint only masked: True, ADetailer inpaint padding: 32, ADetailer version: 23.7.9, Hires upscale: 1.5, Hires steps: 12, Hires upscaler: R-ESRGAN 4x+ Anime6B, Eta: 0.667, Version: v1.4.1
NG(NSFW)
本来这个版本只是我为了自娱自乐制作的,但是由于一些朋友的强烈要求最终决定还是把它上传。请注意,这个版本与前几个版本的质感与内容差别较大,所以并不是所谓的升级版,大家根据自己的需求选择喜欢的版本就好。(另外,这个版本的NSFW内容可能较多,使用请注意)
Originally this version was only made by me for my own amusement, but due to the strong request of some friends finally decided to upload it. Please note that the texture and content of this version differs greatly from the previous versions, so it is not the so-called upgraded version, so just choose the version you like according to your needs. (Also, this version may have more NSFW content, so please be careful when using it)
v1.33
花费了多半天的时间,我找回了最初1.3版本的感觉!不知道发生什么的可以看一下1.3版本的简介,玩得开心!
It took me more than half a day to get back the feeling of the original 1.3 version! Do not know what happened can read the 1.3 version of the introduction, have fun!
v1.3
我犯了一个非常愚蠢的错误,上传错版本了,还把调整好的版本删掉了,而且还没有合成记录,这个版本里的资源是错误的,大家先别下了。。。。。。(太他妈遗憾了,我真的很喜欢这版😭)
I made a very stupid mistake, uploaded the wrong version, but also to adjust the version deleted, and there is no synthesis record, the resources in this version is wrong, you do not download first......(Too damn bad, I really like this version 😭)
v1.2
这个版本和1.1在内容上和效果上几乎没有区别,主要是为了弥补之前融合模型时犯的错误,现在这个版本可以更加准确的识别prompt了。强烈建议停止使用1.1版本。
This version is almost indistinguishable from 1.1 in terms of content and effect, mainly to make up for the mistakes made when fusing models previously, and this version now allows for more accurate recognition of prompts. It is strongly recommended to stop using version 1.1.
v1.1
这个版本增强了渲染质感,也提高了手部表现。但是说实话,能力有限,它的背景风格依然比较单调。这不是第一个版本的完全升级版,它们的风格略有不同。
This version has enhanced rendering textures and improved hand representation. But to be honest, the capabilities are limited and it still has a rather monotonous background style. This is not a complete upgrade of the first version, they have a slightly different style.
这个融合模型可以生成多种多样风格的插画,整体风格偏向动漫写实渲染,对于面部,会根据你的词条改变风格(我个人比较喜欢可爱的风格,所以往里面融了一些大家喜闻乐见的脸模),这个模型对各种词条有还算不错的兼容性,如果你有什么好玩的想法,可以拿它试一试,希望你能享受这个模型带来的乐趣~
我常用的配置是:Steps: 30, Sampler: DPM++ 2M SDE Karras, CFG scale: 20, Seed: 917221560, Size: 512x768, Model hash: 2225758456, Model: POPE2.5_NG4.fp16, Denoising strength: 0.4, Clip skip: 2, RNG: CPU, Hires upscale: 1.5, Hires steps: 10, Hires upscaler: 4x_RealisticRescaler_100000_G, Dynamic thresholding enabled: True, Mimic scale: 8, Separate Feature Channels: True, Scaling Startpoint: MEAN, Variability Measure: AD, Interpolate Phi: 0.7, Threshold percentile: 99.75, Mimic mode: Half Cosine Up, Mimic scale minimum: 3, CFG mode: Half Cosine Up, CFG scale minimum: 3, Eta: 0.667, Version: v1.4.0
Used embeddings: easynegative [119b], bad-hands-5 [10ca]
以下内容为机器翻译/The following is machine translated:
This fusion model can generate a variety of styles of illustration, the overall style is inclined to anime realistic rendering, for the face, will change the style according to your words (I personally prefer a cute style, so into the fusion of some of the popular face models), this model has a pretty good compatibility with a variety of words, if you have any fun ideas, you can take it to try, I hope you can enjoy the fun brought by this model ~
My usual configuration:Steps: 24, Steps: 30, Sampler: DPM++ 2M SDE Karras, CFG scale: 20, Seed: 917221560, Size: 512x768, Model hash: 2225758456, Model: POPE2.5_NG4.fp16, Denoising strength: 0.4, Clip skip: 2, RNG: CPU, Hires upscale: 1.5, Hires steps: 10, Hires upscaler: 4x_RealisticRescaler_100000_G, Dynamic thresholding enabled: True, Mimic scale: 8, Separate Feature Channels: True, Scaling Startpoint: MEAN, Variability Measure: AD, Interpolate Phi: 0.7, Threshold percentile: 99.75, Mimic mode: Half Cosine Up, Mimic scale minimum: 3, CFG mode: Half Cosine Up, CFG scale minimum: 3, Eta: 0.667, Version: v1.4.0
Used embeddings: easynegative [119b], bad-hands-5 [10ca]