CivArchive
    Preview 24958788
    Preview 24958786
    Preview 24958792
    Preview 24958782
    Preview 24958790
    Preview 24958775
    Preview 24958777
    Preview 24958774
    Preview 24958776
    Preview 24958778
    Preview 24958787
    Preview 24958780
    Preview 24958781
    Preview 24958783
    Preview 24958789
    Preview 24958784
    Preview 24958795
    Preview 25092244
    Preview 25097081
    Preview 25098020

    If you want to use more my checkpoint online generation, please visit here.

    https://tensor.art/u/762555264535746522


    V15

    关于例图, 我使用了1024x1360的大小,都是直接输出没有经过AD跟高修.

    For the example images

    V15, I used 1024x1360,

    generated directly without AD or hires fix.

    调整了数据,不在需要正负面质量词.

    The data has been adjusted, and positive/negative quality tags are no needed.

    • Steps: 30

    • CFG scale: 7

    • Sampler: Euler a


    V14

    关于例图, 我使用了1024x1360的大小,都是直接输出没有经过AD跟高修.

    For the example images

    V14, I used 1024x1360,

    generated directly without AD or hires fix.

    重新调整了整体色彩,构图,背景.

    The overall color, composition, and background have been re-adjusted.

    • Steps: 30

    • CFG scale: 7

    • Sampler: Euler a

    Positive Prompt

    score_9, score_8_up, score_7_up,

    Negative Prompt

    score_6,score_5,score_4,, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, sourcwe_pony, source_cartoon,3d, blurry,


    V13

    v13主要是进行了色调跟数据调整,对脸部基本没有太大调整.

    Version 13 mainly focuses on color tone and data adjustments, with minimal changes to the face.

    把cfg稳定在了7,方便后面版本的调整.

    The CFG has been stabilized at 7 to facilitate adjustments in future versions.

    • Steps: 30

    • CFG scale: 7

    • Sampler: Euler a

    Positive Prompt

    score_9, score_8_up, score_7_up,source_real,real skin,
    BREAK

    Negative Prompt

    worst quality,low quality,
    source_furry, source_pony, source_cartoon,2d,
    ,bad anatomy, bad hands,blurry,3d,

    __________________________________________________________________________________________________

    • Steps: 30

    • CFG scale: 5~7

    • Sampler: Euler a

    • Steps: 12

    • CFG scale: 1

    • Sampler: Euler a

    two versions same Prompt

    Positive Prompt

    score_9, score_8_up, score_7_up,source_real,real skin,
    BREAK

    Negative Prompt

    worst quality,low quality,
    worst detail,low detail,
    source_furry, source_pony, source_cartoon,2d,
    ,bad anatomy, bad hands, text, error, missing fingers, (extra digit:1.4), fewer digits, cropped, bad artist, outlines,jpeg artifacts, signature, watermark, username, blurry,3d,

    _______________________________________________________________________

    V11

    • Steps: 30

    • CFG scale: 5~7

    • Sampler: Euler a

    • ADetailer face_yolov8s.pt use (ADetailer sampler: DPM++ 2M, ADetailer scheduler: Karras )

    Positive Prompt

    score_9, score_8_up, score_7_up,source_real,real skin,
    BREAK

    Negative Prompt

    worst quality,low quality,
    worst detail,low detail,
    source_furry, source_pony, source_cartoon,2d,
    ,bad anatomy, bad hands, text, error, missing fingers, (extra digit:1.4), fewer digits, cropped, bad artist, outlines,jpeg artifacts, signature, watermark, username, blurry,3d,

    __________________________________________________________________________________________________

    V10

    • Steps: 30

    • CFG scale: 3.5~7

    • Sampler: Euler a

    • ADetailer face_yolov8n/s.pt use (ADetailer sampler: DPM++ 2M, ADetailer scheduler: Karras )

    Positive Prompt

    score_9, score_8_up, score_7_up,source_real,realistic,photorealistic, 
    BREAK

    Negative Prompt

    worst quality,low quality,
    worst detail,low detail,
    source_furry, source_pony, source_cartoon,2d,
    ,bad anatomy, bad hands, text, error, missing fingers, (extra digit:1.4), fewer digits, cropped, bad artist, outlines,jpeg artifacts, signature, watermark, username, blurry,3d,

    __________________________________________________________________________________________________

    V2toV9

    Recommended Setting

    • Steps: 30-50

    • CFG scale: 7

    • Sampler: Euler a

    • Sampler: Euler a ADetailer face_yolov8n/s.pt use more better

    Positive Prompt

    score_9, score_8_up, score_7_up,

    Negative Prompt

    score_6,score_5,score_4, worst quality, low quality, bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry,

    ________________________________________________________________________________________________

    v1

    Recommended Setting

    • Steps: 30

    • CFG scale: 3.5-7

    • Sampler: Euler a or DPM++ 2M Karras

    • ADetailer face_yolov8n/s.pt use more better

    Positive Prompt

    score_9, score_8_up, score_7_up,

    Negative Prompt

    score_6,score_5,score_4,bad anatomy, bad hands, missing fingers, fewer digits, source_furry, source_pony, source_cartoon,3d, blurry, 

    Description

    FAQ

    Comments (23)

    wyjzpAug 18, 2024
    CivitAI

    好喜欢V10的第六张图片,决定用这张来当SD模型图片

    THEREDONEAug 19, 2024
    CivitAI
    The changes in the model represent the creator's thinking at different stages. In any case, I support you. This is really a very good model.
    Soul_TempleAug 19, 2024
    CivitAI

    经过测试,v10比v9需要更精确的提示词,但是出图效果更好

    lishtai14118Aug 19, 2024
    CivitAI

    请问封面图那种古装的提示词要怎么写?写ancient chinese clothes, traditional chinese clothes, hanfu画出来的都是奇奇怪怪的混搭服饰。

    ============

    更新:看了作者前几期的例图才发现Pony有汉服Lora了,用了Lora后服饰正常多了。

    V10试用下来感觉画面更清晰锐利,妹子也更养眼了,背景细节也更丰富,一些画面特效也能正常表现出来了,算是不小的提升,期待作者下个版本带给我们更多惊喜。

    WAI0731
    Author
    Aug 19, 2024

    请使用古装类的lora 单模型是跑不出很好的古风衣服。我这是早期跑的图使用了古装类的lora 原图丢失了 用的qq群里发的图

    weidwonderAug 19, 2024
    CivitAI

    V10 用Adetailer修复面部不知道为啥, 都垮掉了

    WAI0731
    Author
    Aug 19, 2024

    我的说明文写了,AD需要使用独立的采样

    ADetailer face_yolov8n/s.pt use (ADetailer sampler: DPM++ 2M, ADetailer scheduler: Karras )


    fayerAug 19, 2024
    CivitAI

    pretty good

    daniel691020314Aug 19, 2024· 6 reactions
    CivitAI

    看完大佬的感性宣言,對V10有了點遺憾,這模型小的追求的是最真實的效果,畢竟其他風格在大佬其他模型都涵蓋了

    Soul_TempleAug 20, 2024

    可以自己加提示词:score_9, score_8_up, score_7_up,source_real,realistic,photorealistic,

    Lixioyao20777Aug 23, 2024
    CivitAI

    Version 10 is more cartoon-like in style, and I don't think it's as good as version 9.

    WAI0731
    Author
    Aug 23, 2024

    Version 10 is my first version after transitioning, and I will be adjusting direction in future versions. If you don't like this version, you can temporarily use the older versions of the model. Version 11 is already in development.

    Einstein_Aug 23, 2024
    CivitAI

    good

    saintloris153Aug 23, 2024· 1 reaction
    CivitAI

    与许多服装Lora不通用。试了下载的四十多套衣服,就一两套能用。哪种类型的服装Lora能在这模型里生效?

    WAI0731
    Author
    Aug 23, 2024· 4 reactions

    首先你可以检查一下否使用的是pony的lora,其次pony写实模型本身lora通用性不算很好,下个版本我会提高lora通用性

    saintloris153Aug 24, 2024

    @WAI0731 需要用pony的服装Lora吗?懂了

    yuinyan490363Aug 25, 2024· 3 reactions
    CivitAI

    Why does the higher the weight given to 'long legs', the shorter the socks become, for example, knee-highs turning into ankle socks?

    GGDDGGddAug 27, 2024· 4 reactions
    CivitAI

    歪佬什么时候融一个麦橘7的,隔壁麦橘妹妹好漂亮

    WAI0731
    Author
    Aug 28, 2024

    这个东西怎么说呢,pony本身模型是没有训练过亚洲人所以基本不可能到那个样子的。

    ghost_1252Aug 29, 2024

    @WAI0731 那是不是可以认为,只要当PONY大模型真的融入了足够多的亚洲妹子之后才有希望达到麦橘那样的脸……

    WAI0731
    Author
    Aug 29, 2024

    @ghost_1252 不是那么简单的事情,除非pony作者公开他那个方法,不然训练多了会很僵硬最后就是另一个SDXL.站里面有纯训练的,最后结果就是脸是像了,但是pony灵活性完全没了.

    luguodemanAug 31, 2024

    xsm white pony那个目前是最稳的,看有没有可能融合下

    sarkara486Sep 13, 2024

    麦吉审美疲劳了,可以上libi去看,一堆麦吉的近亲结婚产物,整个站的都一张脸