CivArchive
    veryBadImageNegative - veryBadImageNegative_v1.3
    NSFW
    Preview 283901
    Preview 319375
    Preview 319374
    Preview 319373

    v1.3

    veryBadImageNegative_v1.3使用了新的训练图集，在AOM3和viewer-mix_v1.7中使用时表现较好。

    veryBadImageNegative_v1.3 uses a new training atlas and performs well when used in AOM3 and viewer-mix_v1.7.

    v1.2

    veryBadImageNegative是使用viewer-mix_v1.3生成的特殊图集训练而来的负面嵌入。

    所以veryBadImageNegative是viewer-mix_v1.3的专用负面嵌入。

    或许在其他的扩散模型中也有不错的效果，但缺乏验证。

    veryBadImageNegative is a negative embedding trained from the special atlas generated by viewer-mix_v1.3.

    So veryBadImageNegative is the dedicated negative embedding of viewer-mix_v1.3.

    It may also have a good effect in other diffusion models, but it lacks verification.

    关于v1.2：

    增强图像的质量，削弱了风格。

    在使用v1.2版本时，可以降低一些权重以获得更好的灵活性。

    例如：(veryBadImageNegative_v1.2-6400:0.9)

    About v1.2:

    Enhance image quality and weaken style.

    When using v1.2, you can reduce some weights to obtain better flexibility.

    For example: (veryBadImageNegative_v1.2-6400:0.9)

    关于v1.0与v1.1的差异：

    几乎没有差别，仅用于训练的图集在生成参数有一些微妙的差别。

    关于1600~6400的差异：

    数字代表了训练步数，理论上步数越多效果越好。

    About the difference between v1.0 and v1.1:

    There is almost no difference. The atlas used for training only has some subtle differences in the generation parameters.

    About the difference between 1600 ~ 6400:

    The number represents the number of training steps. In theory, the more steps, the better the effect.

    Description

    FAQ

    Comments (35)

    FluffiesMar 19, 2023· 9 reactions
    CivitAI

    Is this just for anime or also for photorealism and portraits?

    HugoMerinoApr 30, 2023· 3 reactions
    CivitAI

    How do i make this work? I´m using Automatic 1111 Google Colab

    TokilemarSep 11, 2023
    !curl -Lo /content/gdrive/MyDrive/sd/stable-diffusion-webui/embeddings/verybadimagenegative_v1.3.pt https://civitai.com/api/download/models/25820?type=Model&format=PickleTensor&size=full&fp=fp16
    OjisamaMay 1, 2023· 56 reactions
    CivitAI

    can you please upload a safetenors file?

    wiyac33524204May 4, 2023· 39 reactions
    CivitAI

    This is a really good embedding, but could you also make a safetensor variant?

    TheLightningCrabMay 23, 2023· 12 reactions
    CivitAI

    This solved the problems with bad hands really well! Great embedding!

    ishishishiMay 23, 2023· 49 reactions
    CivitAI

    Could you upload a safetensor please?

    lllkojNov 26, 2023

    https://github.com/DiffusionDalmation/pt_to_safetensors_converter_notebook

    fablegeniusNov 29, 2023· 1 reaction

    @lllkoj not that there's any reason to be suspicious but the converter page says "avoid using this notebook to convert a .pt file on your own machine unless that file is from a verified, trusted source. The .pt file is accessed and executed when the notebook is run, along with any potentially malicious code it may contain."

    badgirl1999532Jun 14, 2023· 8 reactions
    CivitAI

    Hello, any body has experience in combining different neg-prompt?

    freezerainJun 15, 2023· 8 reactions

    Hi I was wandering the same and did some tests, you can read here https://civitai.com/models/56519?commentId=158376&modal=commentThread
    in my secon long ass message in comments there are table with examples

    tutu29Jun 15, 2023· 16 reactions
    CivitAI

    大大有国内的爱好者交流社群吗？想加入一起讨论下，感觉自己的使用好像有些问题

    TxbaobaoJul 1, 2023

    有群吗

    Forest_sJul 7, 2023

    蹲

    Satokazima12138Aug 27, 2023

    +1

    wc6974539934Sep 11, 2023

    +1

    wtdy10Oct 28, 2023

    +1

    lyxcom20111Dec 8, 2024

    +1

    zzrdreamJul 29, 2025

    +1

    2622879Sep 29, 2023· 7 reactions
    CivitAI

    It looks like using this inversion comes with texts like something/watermark at the top. Is this supposed to be happend with it?

    angy2218581Dec 25, 2023· 21 reactions
    CivitAI

    PLS, can someone explain to me how to use this? is it a VAE or something? What am I supposed to do? :(

    helperanime460862Dec 25, 2023· 17 reactions

    download this file, and, assuming youre using automatic1111, paste it in this location "G:\StableDiffusion\stable-diffusion-webui\embeddings" (Note: obviously this is my path, but go the location of the embeddings folder in your system). Once done, open a1111, and then, while your cursor is in the second prompt box, go to the tab "Textual Inversion" and click on the name of the file that you downloaded (it should show up, if not, restart).

    angy2218581Dec 25, 2023

    @helperanime460862 ohhh thank you so much <3 but I cannot find the tab "Textual Inversion" :( where is it located?

    KryptonicDec 27, 2023· 4 reactions

    @angy2218581 just below the negative prompt, should have textual inversion, checkpoints, lora, etc

    nanoaiFeb 29, 2024

    Kudos to you for such a good question! 💙

    yofoton174609Mar 6, 2024· 14 reactions
    CivitAI

    Not sure how to use it..

    bakahackerMar 22, 2024· 15 reactions

    From clicking the [!] (Exclamation button):

    Textual Inversions:

    1 - Download the textual inversion

    2 - Place the textual inversion inside the embeddings directory of your AUTOMATIC1111 Web UI instance

    3 - Use the trained keyword in a prompt. In this case the TRIGGER WORD: verybadimagenegative_v1.3 (Used mostly on negative prompt)

    Check image's prompts here for more examples & make awesome images!

    yofoton174609Mar 22, 2024

    @bakahacker thank you. Will the triggerword still work even if I rename the file?

    clarebear8484May 17, 2024· 1 reaction

    @yofoton174609 yes, but you can also use the embedding's file name explicitly. Please look at the many examples provided and the gallery, they show the prompts which should answer your questions. I usually start by copying one of the prompt sets I like and working from there to explore a new embedding/lora/model.

    FrozenGunnerApr 22, 2024· 10 reactions
    CivitAI

    I'm late to the party but dropped in just to say this is great for better hands. Thank you!

    LewdMagicianJul 19, 2024· 9 reactions
    CivitAI

    Sorry Sir. Can I Sell images they generate?

    3973413Aug 6, 2024· 32 reactions

    Yes, but you must give 120% of all profits to owner of this model

    LewdMagicianAug 8, 2024

    OK!? what ?

    Kaz327AIAug 20, 2025· 2 reactions
    CivitAI

    Has anyone encountered malicious code execution due this embedding being a pickletensor file?

    mrusernameiannApr 28, 2026

    I converted this file to .safetensor with Pytorch's safe mode (loads weights_only, disabling any chance of code execution). This, and most others, succeeded. Only somethingsomethingbetterimage_3900 failed to load as weights_only, meaning it's a different structure and may have code lurking inside.

    TextualInversion
    SD 1.5
    by yunleme
    Download (Beta) View on CivitAI
    negative
    negative embedding

    Details

    Downloads
    223,620
    Platform
    CivitAI
    Platform Status
    Available
    Created
    3/19/2023
    Updated
    5/11/2026
    Deleted
    -
    Trigger Words:
    verybadimagenegative_v1.3

    Files

    verybadimagenegative_v1.3.pt

    Size:
    30.92 KB
    SHA256:
    d70463f87042e2b5951c303542f3e171cf49af9b7df53b2f20779493786eb143

    Mirrors

    HuggingFace (145 mirrors)
    CivitAI (1 mirrors)
    TensorFiles (1 mirrors)
    ModelScope CN (1 mirrors)
    TensorArt (1 mirrors)

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.

    SeaArt
    veryBadImageNegative - veryBadImageNegative_v1.3