CivArchive
    Preview 1160610

    Don't let the Model or Lora push a gender on you. Male, female, and every point in between. You decide. This isn't to push any agenda, just giving you the tools to get the image you want, despite the biases that might be in the model you use. Maybe you don't want a dragon with breasts, or you just want to know what Daniel Craig's sister would look like... this will fix that for you.

    I developed this gender slider for use with popular pose and character LORAs that have specific gender training. It makes it hard to get the gender you want when using them. I continued to develop it into a full robust gender slider. These embeddings will fix all gender problems for you. It can make anyone, in any Lora, on any model, any gender. Even animals and fantasy creatures.

    Follow me to make sure you see new tools like this, or new styles, poses and Nobodys when I post them. Things move fast on this site, it's easy to miss.

    How to use:
    Just use one of the positive prompts listed to get the desired results. Use them at various strengths to get the desired results. Make sure they have the appropriate modifier brackets. Example, for automatic1111 it would be (GS-Masculine:0.8), the parenthesis are required. They are stackable, so you can combine them together to help with stubborn models. A negative embedding is also included for extra brute force if needed. Check the version tabs at the top.

    These embeddings push people to the masculine.

    • GS-DeFeminize-Neg: Place in your negative prompt at strength of 1.0 to 1.3, removes beauty, makeup, breasts, and other feminine characteristics.

    • GS-Boyish: Place in your positive prompt at strength of 0.4 to 1.3. Conceptually pushes towards androgynous but more butch or boyish in nature. May vary by model, lora, or prompts.

    • GS-Masculine: Place in your positive prompt at strength of 0.2 to 1.3. Conceptually pushes towards rough and strong seasoned masculinity. May vary by model, lora, or prompts.

    These embeddings push people to the feminine.

    • GS-DeMasculate-Neg: Place in your negative prompt at strength of 1.0 to 1.3. This will removes beards, penises and other masculine features. Great for removing males from a scene.

    • GS-Girlish: Place in your positive prompt at strength of 0.1 to 1.3. Conceptually pushes towards cute and pretty, fun and energetic female in nature. May vary by model, lora, or prompts.

    • GS-Womanly: (coming shortly)

    These are the concepts for the embeddings. Some models, character LoRAs, or embeddings might be too rigid, so increase strength or stack on multiple to get the desired result. Additional prompts may help as well.

    Do you have requests? I've been putting in many more hours lately with this. That's my problem, not yours. But if you'd like to tip me, buy me a beer. Beer encourages me to ignore work and make AI models instead. Tip and make a request. I'll give it a shot if I can. Here at Ko-Fi

    Hide

    Description

    FAQ

    Comments (35)

    MasmalkarJun 15, 2023· 1 reaction
    CivitAI

    Well done my man, as nice as usual, I have some LoRAs that make me curious on what they would look like, as usual thanks!.

    RickMsJun 15, 2023· 1 reaction
    CivitAI

    I just love your work, thank you very much!!!

    SdragonJun 15, 2023· 1 reaction
    CivitAI

    Noticing this have some effect to the clothing as well. Should we remove that concept in the TI to allow of more feasible character design?

    ThomboJun 15, 2023

    Also seeing that It has a very strong bias for certain types of clothing

    Zovya
    Author
    Jun 15, 2023· 1 reaction

    the TI doesn't use any tokens regarding clothing. models will have a bias to different ages and genders however. just prompt the clothing you like

    SerenityXJun 16, 2023
    CivitAI

    Got a sample image for the Womanly variant?

    104450Jun 17, 2023· 7 reactions
    CivitAI

    For naming nomenclature consistency, wouldn't it be better for either the "Masculine" slider to be renamed "Manly" to match the "Womanly" slider, or the "Womanly" slider be renamed "Feminine" to match the "Masculine" slider?

    Zovya
    Author
    Jun 18, 2023

    Good idea. Also the great thing about TIs is you can name them whatever you like. Just use the filename you give it as the trigger word in the prompt.

    anon4324Jun 18, 2023· 6 reactions
    CivitAI

    Works amazingly well

    Call me a skeptic, but I really didn't think this would work as well as it did. I really had my doubts that this would be able to take prompts and Loras of characters that are definitely feminine/masculine and swap them around so effortlessly. Not only that, but actually keeping close to the character design. I was even more skeptical when I saw it was a Textual Inversion, and not a Lora. I seriously wrote of TI's as obsolete compared to Lora, but this is an amazing example of what it can do in the right hands.

    Depending on your prompts (and whether or not they have distinctly masculine/feminine genitals), you might have a dude with the opposite undercarriage or vice-versa, but if I know the internet, there's plenty of people into that thing, so it's not a bug, it's a feature!

    I've seen some comments that clothing can be significantly altered depending on things, but as I largely prompt characters sans-clothes, I can't add any input on that front.

    Still, it's an incredible set of embedding to have in one's arsenal, and the Age Slider is another one to be sure! Definitely keep an eye on this user's stuff in the future.

    Zovya
    Author
    Jun 18, 2023

    thanks so much for that

    anon4324Jun 19, 2023· 2 reactions

    @Zovya No problemo! I've had it and the age slider downloaded for a while now, but I finally gave them a curious go, and I was really impressed by how well it worked. The fact that the character designs stay so faithful to the original is the same kind of magic as when you prompt an image and see it come out for the very first time. Like seriously, pop on one of the gender swaps, hit the generate and it's like "yup, that's exactly how I imagined it". So many different conceptual things fall flat, but this one actually does what it says on the tin, no fuss no muss. Great work.
    I can't wait to see what you come out with next!

    CaL9Mi7YJun 25, 2023· 12 reactions
    CivitAI

    Whether by birth or environment, I've always had a strong streak of extremely fierce independence, and there's a reason why I love your "Slider" series of tools. They allow me to NEVER let ANY Faction dictate their age, race AND gender "preferences" to me. Live and let live. Some might argue you still haven't done weight or beauty sliders--if that's even possible, but I think you got the essentials covered. At least for me.

    Thank you for the great job as always, author-san. Have an awesome week ahead, cheers. ❤

    Zovya
    Author
    Jun 27, 2023· 1 reaction

    thanks. those are also some good ideas

    hcommand68990Jul 8, 2023· 17 reactions
    CivitAI

    I fully support Stallone coming out as trans. It was brave.

    cherryblossomAIJul 8, 2023
    CivitAI

    I am new to stable diffusion, can anyone tell me how to use this in automatic1111 webui please?

    misterandy33Jul 8, 2023· 1 reaction

    Download. Put into "webui\embeddings" folder. Put trigger words into prompt.
    https://github.com/civitai/civitai/wiki/How-to-use-models#textual-inversions

    Note that this is a Textual Inversion. Loras go into "webui\models\Lora"; checkpoints go into "webui\models\Stable-diffusion"; etc. See the link for more info.

    cherryblossomAIJul 9, 2023

    @misterandy33 thanks you, I just did as you mentioned, now my question is can I apply it in my img2img prompt as well? I am applying the designated prompt in the img2img prompt box but not getting the desired result. specifically for the female to male cases, where the image should look almost the same as the former image but with more masculine features. Also my question is what should I set the denoising level to if I use it on img2img? Thank you for replying.

    misterandy33Jul 10, 2023

    @cherryblossomAI I wish I could help with that, but I haven't done any img2img stuff yet... :(

    cherryblossomAIJul 10, 2023

    @misterandy33 Ahhh I see, maybe I will need to try out some tweaking with the denoising slider then.

    cherryblossomAIJul 13, 2023
    CivitAI

    any help on how to use it in img2img? I have tried hundreds of times, but it seems it can't push the female body or face to a male equivalent, same can't be said for female to male scenarios though. Any ideas are welcome.

    Zovya
    Author
    Jul 14, 2023

    increase denoising strength

    cherryblossomAIJul 15, 2023· 3 reactions

    @Zovya Hi, I would like to share some of my experiences, I have tried by increasing the denoising strength but it would change the whole scene and background, so I decided to use a different technique instead and it worked a lot better with this prompt.

    1. Firstly I set the denoising strength to around .1

    2. Disable control net, as it's completely irrelevant an will get into the way of the prompts.

    3. stack the prompts for more strength if needed.

    4. using Euler A as sampler and sampling setting sampling steps to 18-25.

    5. Most important set CFG scale as high as possible to stop generating creative & make it stick to the prompt heavily to get the best manly feature on even the most feminine body.

    That's it that's all and hopefully this will be able to make any feminine character with huge bosom into a big chested manly man.

    LazmanAug 28, 2024

    @cherryblossomAI Are those all options in the embedding, or in the program itself? A lot of people asking these questions aren't mentioning the software they use, and those answering aren't mentioning which software they refer to (if software is relevant). So, it can all get a bit confusing. Personally I use Easydiffusion, since it's free, simple to use, and I can download it for offline use --Which I personally prefer, because I'm old-school, and still carry this odd notion that we should not have everything we do surveilled by corporations.

    JanetJul 21, 2023· 3 reactions
    CivitAI

    I wonder if it's possible to improve on, or maybe make a variant of, the boyish slider. All other sliders work very well, but I suspect this one's difficult because SD has a huge bias to young = feminine. I notice boy characters very easily develop long hair, jewelry, feminine eyes and jawlines. I use the de-feminine negative, and the boyish positive, but too strongly and it'll give stubble, not strongly enough and he'll still have girlish eyes/jaw/hair.

    LazmanAug 29, 2024

    I feel your pain.. Just be lucky if your boys aren't randomly produced with breasts..

    kodybrown125Aug 8, 2023· 11 reactions
    CivitAI

    Stallone is handsome as a man and beautiful as a woman! ;-)

    hushichoDec 21, 2023· 2 reactions
    CivitAI

    I'm so glad to finally find this, because it's an invaluable resource. It helps appreciably and palpably.

    Themis_PMar 21, 2024
    CivitAI

    This embedding has been incredibly helpful. I really appreciate it. Thanks a lot.

    garifetdinov604Jul 14, 2024· 5 reactions
    CivitAI

    Is it possible to do the same for SDXL or PDXL? That would be excellent. Thank you for your work!

    LazmanAug 29, 2024· 6 reactions
    CivitAI

    Let's be real, this add-on exists so we can reliably produce art of males without tits. I'm not hating on women, or female parts. Just sayin, it's jarring to see Killua or first season Naruto with boobs.. There's a weird systemic bias when it comes to inserting those into everything..

    darkstar82Sep 27, 2024· 2 reactions

    99% of models are produced by straight men between the age of 20 and 40 years old... they tell a psychological story alllll of their own lol.

    LazmanJan 23, 2025

    @darkstar82 I am a man, 44 years old, and tbh, even I am disgusted/appalled by all these grown men that act like mentally handicapped toddlers when it comes to seeing naked women.. It's like they just start bouncing up and down in their seat while clapping and going "Dur... tee hee hee hee, boobies!!!"

    That's not all men, obviously, I'm not, but far too many. And much like alcohol glorification, that behavior is largely propagated by the entertainment media and advertising.

    I know you're thinking it's because Hollywood is a patriarchy, and while such claims may not be entirely false when it comes to studio practices and biases, I do believe the push to keep people stupid and divided goes a good bit deeper than 'me man, you woman, we fight/screw now'.

    Edit, just thought of something.. One caveat to your response. You may be correct about those that produce the models, but what about those using them? Prob the majority using them are just paying for corporate site use, so that could be literally anyone, yet the vast majority of the nude imagery I see is of women. That's one thing I've never got, women like dicks, right? And it's not like they have to hide that while on the net, so why is there so little demand for dicks? Or are all women secretly lesbians(jk)?

    4314188Sep 8, 2024
    CivitAI

    How do I use this?!

    OverlonerOct 26, 2024
    CivitAI

    wow

    cognMar 24, 2026
    CivitAI

    is there any chance of getting these as safetensors? thanks either way :)

    TextualInversion
    SD 1.5

    Details

    Downloads
    10,304
    Platform
    CivitAI
    Platform Status
    Available
    Created
    6/15/2023
    Updated
    5/13/2026
    Deleted
    -
    Trigger Words:
    GS-Womanly

    Available On (2 platforms)

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