--
如果你想在我的模型上發布你的作品,可以一次在同一個post發布20張圖片,如果是優秀的作品我就能將其頂至在展示圖片的最上方。
"If you want to post your work on my model, you can upload up to 20 images in a single post. If the work is outstanding, I can feature it at the top of the displayed images."
--
模型介绍为中英双语,中文用户可以关闭翻译。
--
显示卡低配备者建议使用”Hyper-sd1.5、sdxl的lora加速器。
下载Hyper-SD15-8steps-lora或Hyper-SD15-12steps-CFG-lora.
取样方法(Sampler)=Euler a.
Schedule type: SGM Uniform
CFG:5~6
--
Those with low graphics cards are recommended to use the "Hyper-sd1.5 or sdxl" lora accelerator.
Download Hyper-SD15-8steps-lora or Hyper-SD15-12steps-CFG-lora.
Sampler=Euler a.
Schedule type: SGM Uniform
CFG:5~6
---
如果你喜欢我的模型,可以用"buzz"鼓励我。
If you like my model, you can encourage me with "buzz".
--
我只有在” civitai”发布我的模型和lora,其余的地方皆是盗用。
I only publish my models and lora on ” civitai” , the rest are all stolen.
--
客製化lora:我的Telegram ID:@iamddtla0620
我的微信ID:iamddtla
我的line ID:iamddtla
时间宝贵,请直接说明来意,例如:订制LORA.
Customized lora: My Telegram ID: @iamddtla0620
My WeChat ID: iamddtla
My Line ID: iamddtla
Time is precious, please state your intention directly, for example: customized LORA.
--
面部局部重绘可使用adetailer或sd-face-editor.
使用"sd webui cutoff"可避免颜色溢出污染其它元素。
高解析度放大可使用"ultimate SD upscale".
--
For partial facial redrawing,adetailer orsd-face-editor can be used.
Use "sd webui cutoff" to avoid color overflow contaminating other elements.
High resolution upscaling can be done using "ultimate SD upscale".
2024/10/06
V6:
1.完全的重新合并,合并的模型皆为”培训(Checkpoint Trained)”的模型,并免合并混和”(CHECKPOINT MERGE)”的模型。没有任何lora的合并。
2.模型挑选高品质模型、高解析度的模型、或是具有大量培训数据的模型。
3.更好的理解自然语言与danbooru标签。
4.模型不包含vae.
5.你可以尝试生成1024 解析度的图片。 在1024解析度时,将更容易生成良好手指。要生成站立的全身图片,也建议生成高度为1024解析度的图片.
6.模型不包括 VAE,如果没使用VAE生成的图片会较灰暗,推荐"vae-ft-mse-840000-ema-pruned"(颜色较鲜艳)或orangemix.vae (颜色较浅)。
--
1. Complete re-merge, all merged models are "Checkpoint Trained" models, and no "CHECKPOINT MERGE" models are merged.There is no merging of any lora.
2. Model selection: high-quality models, high-resolution models, or models with a large amount of training data.
3. Better understand natural language and danbooru tags.
4.The model does not contain vae.
5.You can try to generate the image at 1024 resolution. At 1024 resolution, it will be easier to generate good fingers. To generate standing full body images, it is also recommended to generate highly 1024 resolution images.
6.The model does not include VAE. If VAE is not used, the image generated will be darker. We recommend "vae-ft-mse-840000-ema-pruned" (brighter color) or orangemix.vae (lighter color).