QQ|Archiver|手机版|小黑屋|AiOMG ( 粤ICP备2022147020号-1 )

GMT+8, 2024-10-16 20:22 , Processed in 0.039205 second(s), 29 queries .

Powered by Discuz! X3.5 Licensed

© 2001-2024 Discuz! Team.

[LoRA] Gundam RX78-2 outfit style 高达RX78-2外观风格

21
0
0
0
simonlive LV9
正文
发布时间:2024-07-27
模型类型
模型类型: LoRA
基础模型: SD 1.5
最新版本: SD1.5 V1.0
触发词: gundam\(rx78\), mecha, robot, science fiction
下载链接: https://pan.baidu.com/s/13ga_0gOyaHQqIISKN-0VAQ?pwd=fr6j
提取码: fr6j
源链接: https://civitai.com/models/22470?modelVersionId=26830
模型简介:
Latest update: SDXL version 2.0 released
2023/12/09

Improved TE training, Great common use ability now.
Can perform really good with SDXL base 1.0, White Base, please make a try
take the prompt from my images.
use the weight from 0.5-0.8
it is very like SD1.5 time, it's a Gundam cloth for Girl!
make your Gundam Girl now!
again:

~~~~~~~~~~~~~~~~~~~~~~~o(* ̄︶ ̄*)o~~~~~~~~~~~~~~~~~~~~~~~

Face editor Plugin is highly recommended
Latest update: SDXL version released
Instruction: please notice SDXL version is a Lycoris not Lora, but same way to use

This is the updated version for SDXL, if you wish to try SDXL, you can download this one, and it is a Lycoris version, for some reason, SDXL base model has a strange weight for Gundam, once you have the girl face, it will not likely to show full body, please use high weight to trigger it, or adding more prompts which related to the leg, foot, such as, long leg, shoes, it will help to full body prompt to realize.

if you find your output is likely to only have underwear, please use prompt bodysuit, it will significantly help to put the cloth on.
finally, please take a try and discover more!

注意:SDXL版本实际上是Lycoris版本,而不是Lora版本,但使用方法相同。

这是SDXL的更新版本。如果您想尝试SDXL,可以下载这个版本,它是Lycoris版本。由于某种原因,SDXL基础模型对Gundam的权重有些奇怪,一旦您得到女孩的脸部,它不太可能显示全身,请使用高权重来触发它,或者添加更多与腿、脚相关的提示,例如,长腿、鞋子,这将有助于全身提示的实现。

如果您发现您的输出只有内衣,请使用提示词“bodysuit”,它将显著地帮助添加衣物。

最后,请尝试并发现更多!

~~~~~~~~~~~~~~~~~~~~~~~o(* ̄︶ ̄*)o~~~~~~~~~~~~~~~~~~~~~~~

Face editor Plugin is highly recommended

欢迎查看我的其他模型:ttplanet Creator Profile | Civitai

这是一个非常平衡的lora, 可以恰如其分的展现高达的特征,但是却又可以比较好的控制权重,使用gundam\(rx78\)作为主触发词放在prompt的第一个,然后根据你使用的模型适当的调节权重从0.1-1都可以。

你可以让真人穿上高达风格的衣服,也可以通过mecha armor触发关键词直接穿上盔甲,这取决于你需要什么风格。

这个lora可以很好的和各种模型匹配,不管是真人的还是动画的!而且可以和各种其他人物lora混用!
建议使用方法是先正常的生成一个人物,调整到你喜欢的衣服和动作,调整你喜欢的场景,然后加载Lora, 如果你使用的是系统lora插件,那就直接选用添加权重,如果你不使用插件,那么就用<lora:gundamRX782OutfitStyle_v10.safetensors:1>进行触发,调整权重。

请查看我的范例图,里面是一些想法,你们可以自己玩的更high。

lora使用512*512采样生成,理论上768以下的解析度都能较好支持,建议使用hiresfix进行高清修复。想到什么再更新吧,希望能得到你们的支持。

推荐使用底模recommend SD model:
chilloutmix
Daceflopainting
AOM系列
Anything系列

以上模型都可以涩涩,如果你喜欢机甲瑟瑟的话。

This is a very balanced Lora that can perfectly showcase the features of Gundam while also allowing for good control over the weighting, use gundam\(rx78\) as the priority prompt, ranging from 0.1 to 1 depending on the model you're using.

You can have a real person wear Gundam-style clothing, or trigger keywords to directly wear mecha armor, depending on the style you prefer.

It fit will most SD model no matter it is for real person or animation.
The recommended method is to first generate a character normally, adjust their clothing and movements to your liking, adjust the scene to your liking, and then load Lora. If you're using the system Lora plugin, you can directly add weighting. If you're not using the plugin, you can trigger it with <lora:gundamRX782OutfitStyle_v10.safetensors:1> and adjust the weighting.

Please refer to my sample images for some ideas, and feel free to explore on your own.

Lora is generated at 512*512 resolution, and can support resolutions below 768 fairly well. We recommend using hiresfix for high-definition restoration. We'll update as more ideas come up, and I hope to have your support.

~~~~~~~~~~~~~~~~~~~~~~~o(* ̄︶ ̄*)o~~~~~~~~~~~~~~~~~~~~~~~

示例提示词和参数:
Prompt: mini\(ttp\), gundam\(rx78\), (8k, RAW photo, best quality, masterpiece:1.2),solo,cybertron,RX-78-2, (full body), mask, dancing, (isometric), miniature,<lora:miniature_V1:0.7>, ((simple background))

Negative prompt: (((text))),(((logo))),(((watermark))),(((sea))),(((blur))), (EasyNegative:1.2), ng_deepnegative_v1_75t, paintings, sketches, (worst quality:2), (low quality:2), (normal quality:2), lowres, normal quality, ((monochrome)), ((grayscale)), bad anatomy,(long hair:1.4),DeepNegative,(fat:1.2),facing away, looking away,tilted head,lowres,bad anatomy,bad hands, text, error, missing fingers,extra digit, fewer digits, cropped, worstquality, low quality, normal quality,jpegartifacts,signature, watermark, username,blurry,bad feet,cropped,poorly drawn hands,poorly drawn face,mutation,deformed,worst quality,low quality,normal quality,jpeg artifacts,signature,watermark,extra fingers,fewer digits,extra limbs,extra arms,extra legs,malformed limbs,fused fingers,too many fingers,long neck,cross-eyed,mutated hands,polar lowres,bad body,bad proportions,gross proportions,text,error,missing fingers,missing arms,missing legs,extra digi

Steps: 40, CFG scale: 9, Sampler: Euler a, Seed: 2880182193, ENSD: 31337, Size: 640x1152, Model: revAnimated_v11, Model hash: d725be5d18, Hires steps: 30, Hires upscale: 2, AddNet Enabled: True, AddNet Model 1: Skeleton_01(e351fd36361b), AddNet Model 2: RX78_V1(c477ce14de20), Hires upscaler: ESRGAN_4x, AddNet Module 1: LoRA, AddNet Module 2: LoRA, AddNet Weight A 1: 0.9, AddNet Weight A 2: 0.5, AddNet Weight B 1: 0.9, AddNet Weight B 2: 0.5, Denoising strength: 0.3, Clip skip: 2

本帖子中包含更多资源

您需要 登录 才可以下载或查看,没有帐号?立即注册 微信登录 手机动态码快速登录

x
回复

使用道具

 
您需要登录后才可以回帖 登录 | 立即注册 微信登录 手机动态码快速登录

本版积分规则

快速回复 返回顶部 返回列表