RealMixPony

请详细查看说明！Important, please read the Description before use.=======================================================RE v0.3Adjust colors, make it more cinematic, and make characters more realisticRE v0.2Fix issues, improve colorRE v0.1Reduced some realism, will look slightly more 3D than before, improved faces and backgrounds=======================================================I don't speak English, so please excuse me if the translation is wrong.此模型从DucHaiten-Pony 和 real pony合并而来.请支持原作者.支持写真和XXX类型图.建议用Comfyui，工作流如下：https://civitai.com/articles/5579注意默认脸看起来像未成年，需要添加年龄提示词.全身图片脸部会崩坏，这个是因为脸部分辨率不足引起，请使用FaceDetailer或者ADetailer修复。A Merge of DucHaiten-Pony and real pony.Please support the original author of the model.Realistic portrait photography style,Pony support.Can be used on Webui and ComfyuiIf you are not familiar with Comfyui, Download Comfyui WorkFlow here:https://civitai.com/articles/5579Note that the default face looks like a minor, so you need to add age prompts.The face of the full-body image will be corrupted. This is caused by insufficient face resolution. Please use FaceDetailer or ADetailer to fix it.============================================================当动态cfg关闭时:When Dynamic Thresholding is disabled：steps:25cfg:5sampler:euler a karrasresolution:1024x1024当动态cfg打开时:When Dynamic Thresholding is enabled：steps:25cfg:20sampler:euler karrasresolution:1024x1024写提示词，建议按我的模板（删除红色字，写入你自己需要的）Suggested prompt words，Replace the red words with your ownPrompt:score_9,score_8_up,score_7_up,拍摄角度(full_body, half_body, medium_full_shot,等)Shooting angle (full_body, half_body, medium_full_shot, etc.)1girl,adults,(mature:0.8),perfect_face,thin_face,你需要的提示词 (narrow waist, sitting, reading, library,等)The prompt words you need (short_hair,narrow waist, sitting, reading, library, etc.)sexy_view,unique_pose,unique_perspective,wide_view,Expressiveh,volumetric_lighting,vivid_colours,rembrandt_lighting,blurry_background,depth_of_field,high_contrast,dramatic_shadows,dramatic_lighting,highly_detailed,high_contrast,film_grain,Negative prompt：score_4,score_5,score_6,newborn,teen,child,toddler,infant,cherub,college_age,middle_age,grandmother,grandfather,mother,mature_female,milf,worst_quality,low_quality,normal_quality,messy_drawing,amateur_drawing,low_resolution,bad_anatomy,bad_hands,source_furry,source_pony,source_cartoon,comic,source_filmmaker,3d,showing,teeth,censor,bar_censor,mosaic_censorship,============================================================以下为Comfyui教学The following is Comfyui tutorial1.使用SDXL风格化提示词,指定你需要的风格，也可以不使用，看你需求。Use SDXL Prompt Styler,Specify the style you want，You can also close it, depending on your needs2.Low Light Lora(optional)低光lora，更能突出真实性，看自己需求。https://civitai.com/models/623662/low-light-for-pony-variantstrength_model:0.6-0.8strength_clip:0.6-0.8Lora Block Weight：1,1,1,0,0,0,0,0,0,0,0,03.使用动态CFGDynamic Thresholding(Advanced)https://github.com/mcmonkeyprojects/sd-dynamic-thresholding?tab=readme-ov-file设置如下图:The settings are as shown below:当使用动态cfg时，ksampler设置如下图When using Dynamic Thresholding, Ksampler is set as follows:4.使用FaceDetailer个人不建议使用，除非在浅空间放大后脸部效果还是不行。I personally don't recommend using FaceDetailer unless the face is still bad after using Latent Upscale.FaceDetailer settingsModel:face_yolov8n_v2.ptSAM Model:sam_vit_h_4b8939.pthSegm_Detector:face_yolov8n-seg2_60.ptsteps:20-25，cfg:5（When using Dynamic Thresholding, CFG is set to 20）sampler:euler a karras（When using Dynamic Thresholding,Sampler is set to euler karras）denoise:0.4-0.6============================================================以下是给初学者，你熟悉操作的话不用看The following is for beginners. If you are familiar with the operation, you don’t need to read on.Simple picture drawing process:1.Enter the prompt word you need，Replace the red words with your own写提示词，建议按我的模板（删除红色字，写入你自己需要的）Prompt:score_9,score_8_up,score_7_up,拍摄角度(full_body, half_body, medium_full_shot,等)Shooting angle (full_body, half_body, medium_full_shot, etc.)1girl,adults,(mature:0.8),perfect_face,thin_face,你需要的提示词 (narrow waist, sitting, reading, library,等)The prompt words you need (short_hair,narrow waist, sitting, reading, library, etc.)sexy_view,unique_pose,unique_perspective,wide_view,Expressiveh,volumetric_lighting,vivid_colours,rembrandt_lighting,blurry_background,depth_of_field,high_contrast,dramatic_shadows,dramatic_lighting,highly_detailed,high_contrast,film_grain,Negative prompt:score_4,score_5,score_6,newborn,teen,child,toddler,infant,cherub,college_age,middle_age,grandmother,grandfather,mother,mature_female,milf,worst_quality,low_quality,normal_quality,messy_drawing,amateur_drawing,low_resolution,bad_anatomy,bad_hands,source_furry,source_pony,source_cartoon,comic,source_filmmaker,3d,showing,teeth,censor,bar_censor,mosaic_censorship,For example：Prompt:score_9,score_8_up,score_7_up,medium_full_shot,1girl,adults,(mature:0.8),perfect_face,thin_face,medium_breasts, narrow_waist, camisole, short hair, tattoos, hands in pockets, disdain, street, trash can, graffiti,sexy_view,unique_pose,unique_perspective,wide_view,Expressiveh,volumetric_lighting,vivid_colours,rembrandt_lighting,blurry_background,depth_of_field,high_contrast,dramatic_shadows,dramatic_lighting,highly_detailed,high_contrast,film_grain,Negative prompt:score_4,score_5,score_6,newborn,teen,child,toddler,infant,cherub,college_age,middle_age,grandmother,grandfather,mother,mature_female,milf,worst_quality,low_quality,normal_quality,messy_drawing,amateur_drawing,low_resolution,bad_anatomy,bad_hands,source_furry,source_pony,source_cartoon,comic,source_filmmaker,3d,showing,teeth,censor,bar_censor,mosaic_censorship,2.Select the SDXL Prompt Styler you need，or base加入SDXL风格化提示词，你也可以默认3.Run the workflow and your image will start to be generated.运行工作流，出图。Result (for example)4.On Upscale Workflow, load your favorite image在放大工作流中，上传你觉得不错的图5. copy the prompt words从出图工作流拷贝提示词到放大工作流。6.If desired, you can enable low-light Lora and Vivid Midjourney mimic Lora如果需要，启用低光lora和Midjourney模拟lora，记得写上lora block weight。https://civitai.com/models/623662/low-light-for-pony-varianthttps://civitai.com/models/595237?modelVersionId=665272Enter the strength you want，Remember to enter the Lora Block Weight value.7.I personally don't recommend using FaceDetailer unless the face is still bad after using Latent Upscale.个人不建议使用，除非在浅空间放大后脸部效果还是不行。8.Compare the enlarged image with the original image。图片比较节点，查看放大后图片效果。9.The final generated example最后出的例图。