CivArchive
    Preview undefined

    Buy me coffee
    https://unfazedanomaly964.gumroad.com/l/buymeacoffee

    Best result with weight between : 0.5-0.8 and the CFG between 6-8
    Reviews are really appreciated, i love to see the community use my work.
    If you like my work, please tip me

    there are some issues related with hands. Next version will most likely be about that,

    Description

    FAQ

    Comments (52)

    CharlieBrown0115Jul 25, 2025· 3 reactions
    CivitAI

    looks nice , wish you can make a lora of big dic..s slaping big fat @sses, lol

    unfazedanomaly964
    Author
    Jul 26, 2025

    I’ll give it a shot, but no promises.

    lucky_chanJul 25, 2025
    CivitAI

    Awesome...

    unfazedanomaly964
    Author
    Jul 26, 2025

    thanks

    docilerJul 26, 2025· 1 reaction
    CivitAI

    I like the 'fingers pushing into the skin' kind of stuff, still looking how to describe that better. Good job on that!

    unfazedanomaly964
    Author
    Jul 27, 2025

    thanks

    SantaonholidaysJul 28, 2025· 2 reactions
    CivitAI

    Save the Training Data for 2.2 :)

    GoesNowhereJul 29, 2025· 1 reaction
    CivitAI

    such a good work!
    but I would like to see more about "fingers pushing into the skin" effect,for now it seems just two hand moving with no pressure on breasts.
    can we add "nipple pull" into lora?

    unfazedanomaly964
    Author
    Jul 29, 2025

    There is a huge list of commission work I need to do. I am not sure when I can even start to collect dataset for that

    unfazedanomaly964
    Author
    Jul 29, 2025

    But I will consider it when I can

    GoesNowhereJul 29, 2025

    unfazedanomaly964 thank you,hope you have a nice day

    gumpbubba721291Jul 30, 2025· 1 reaction
    CivitAI

    Ok so from my experience with using it for some time today, I would say it's a great lora. The strength it has I would say over other similar wan lora is that it handles the skin indentation I think pretty naturally looking. I imagine that's difficult to get right, so big congrats on that. It gets an actual grab motion down well and handles the sort of pull of flesh, if that makes sense.

    Something I would like to see in the future on a fondle type lora is more variation in speed. I haven't seen any of the fondling type lora handle fast grabs or sort of juggling (jiggling?) breasts. This lora has one speed though it does it well. I feel like that's probably the biggest weakness it has, though it's not like there is a lora that has yet really nailed that from what I've seen. All in all, great job! Thanks for sharing!

    unfazedanomaly964
    Author
    Jul 30, 2025

    thanks for the feedback

    bluenightlagoonAug 3, 2025
    CivitAI

    Awesome Model!!! Wish I could support you more. But I can't buy buzz anymore. All options are down.

    unfazedanomaly964
    Author
    Aug 3, 2025

    Thank you! Even just taking the time to leave a comment means a lot. I really appreciate it!

    unfazedanomaly964
    Author
    Aug 7, 2025

    If you'd still like to support me, you can buy me a coffee here:
    https://unfazedanomaly964.gumroad.com/l/buymeacoffee

    CrystalVisageAug 5, 2025
    CivitAI

    Amazing lora. Thank you for making it. I tested with strength 1, and it makes her feel almost like jelly under the fondling hand. Strength 0.6 was better, but still it's a bit too "fluid". That said, it's very good. Hopefully we'll get a more natural looking fondling at some point?

    Need firmer tits!!

    Thanks regardless.

    96857Aug 5, 2025
    CivitAI

    Is there a t2v for 2.1 and 2.2 planned?

    unfazedanomaly964
    Author
    Aug 5, 2025· 1 reaction

    I don’t have any set plans at the moment. Most of what I do is spontaneous and inspired by whatever I come across online. Right now, I can’t make any promises—I’m heading into a very busy period and will likely be away from home for a large part of the year. I’ll try to work on it when I can, but I can’t guarantee anything.

    96857Aug 5, 2025

    unfazedanomaly964 could you share the dataset somewhere?

    unfazedanomaly964
    Author
    Aug 5, 2025· 2 reactions

    chezolver I don’t keep datasets for long after training. I erase them to free up space for the next training. This one was trained 11–12 days ago and has already been deleted.

    recilurnAug 5, 2025· 1 reaction
    CivitAI

    GOAT creator

    unfazedanomaly964
    Author
    Aug 5, 2025

    What is a goat creator? Can you give more description?

    HQdesuAug 8, 2025

    unfazedanomaly964 Greatest Of All Time=GOAT

    unfazedanomaly964
    Author
    Aug 8, 2025· 1 reaction

    HQdesu thanks. I am a rookie and still learning.

    ClocksmithAug 6, 2025· 2 reactions
    CivitAI

    This has categorically stolen the crown from the exotic breast massage lora. I was fighting with that thing for 30 generations and could barely get her to make feather touches to her breasts. This one had full on grasping in one gen.

    MegaWalrusAug 6, 2025

    I agree, this is excellent. For less deep 'grasping' and more just heavy self-caressing, the Exotic Dancer LoRA also seems to work well when no 'dance-y' prompts are used. Both have good 'Squish Physics'

    ainsfwloveAug 9, 2025· 5 reactions
    CivitAI

    LoRA works perfectly with my t2v on Wan 2.2 — perfect! No need for images, just text is enough.

    trollblasterAug 9, 2025· 2 reactions
    CivitAI

    anyone can give me some advise? i want to use some of these awesome wan models, but im not sure what i need, im using a1111 right now, is there some guide i can see? anyone can help me? thanks in advance.

    unfazedanomaly964
    Author
    Aug 10, 2025

    use comfyUI or CivitAI generator. But CivitAI generator is not that good. There are some good workflows on CivitAI for comfyUI wan video generation.

    trollblasterAug 10, 2025

    unfazedanomaly964 thanks... damn i cant use comfyui, it gives me headache...

    unfazedanomaly964
    Author
    Aug 10, 2025

    trollblaster It is the best way if you have good gpu

    StableNerdAug 12, 2025· 1 reaction

    You can use Pinokio app locally. look up on YouTube, how to install it (it's very easy), then you can install wan 2.1/2.2 within the app. Watch the videos about it first, then decide if it's for you :)

    trollblasterAug 13, 2025· 1 reaction

    StableNerd thanks, i try that.

    Styl1shAug 27, 2025· 2 reactions

    I'm having a good time with Wan2GP, it's available on Pinokio, or I can help you with regular setup

    Dim796Aug 12, 2025· 5 reactions
    CivitAI

    Awesome

    lendastaAug 16, 2025· 2 reactions
    CivitAI

    Is this for Wan 1 or Wan 2? And.. Wan 1 lora could be used for 2? Thanks! :)

    unfazedanomaly964
    Author
    Aug 16, 2025· 3 reactions

    it is wan 2.1. Yes wan 2.1 can be used for low noise wan 2.2

    lendastaAug 18, 2025· 2 reactions

    unfazedanomaly964 and what you put in high noise? You use only one lora in low noise? Thanks for the answer :)

    unfazedanomaly964
    Author
    Aug 18, 2025· 3 reactions

    lendasta WAN 2.1 LoRAs usually work fine in WAN 2.2 low-noise since the setup is very similar. For high-noise, though, you generally need a LoRA specifically trained for WAN 2.2.

    lendastaSep 11, 2025

    @unfazedanomaly964 thanks!

    darkroast175696Sep 30, 2025· 2 reactions
    CivitAI

    The model is amazing and I'm only using it for wan 2.2 text-to-video. My only complaint isn't even the model's fault: civitai won't let me link the model in combination with the other loras I'm using because this model is listed as image-to-video. If I could reference it, I'd list this lora on most of my videos lately. Great work!

    zirtapodOct 21, 2025
    CivitAI

    I'm using this for I2V high or low?

    unfazedanomaly964
    Author
    Oct 22, 2025

    This is Wan 2.1 so you can use it as low for Wan2.2

    unfazedanomaly964
    Author
    Jan 20, 2026

    I made a Wan 2.2 version. Please give it a shot.

    travahakaJan 1, 2026
    CivitAI

    Thank you! Hoping for an updated 2.2 version...

    UncleJertJan 1, 2026

    I second this. This lora is bumpin' and I've hoped for a 2.2 version since 2.2 became a thing.

    unfazedanomaly964
    Author
    Jan 1, 2026

    @UncleJert My laptop crashed and I lost the dataset for this training. I tried to train with a new dataset but it was not as good as this. I will try it again later

    UncleJertJan 1, 2026

    @unfazedanomaly964 All good. Shit happens! Looking forward to it.

    unfazedanomaly964
    Author
    Jan 20, 2026· 2 reactions

    I have just uploada 2.2 version. Please give it a try

    UncleJertJan 21, 2026

    @unfazedanomaly964 My man!! 👍

    travahakaApr 10, 2026· 2 reactions

    Thank you so much! Will give it a spin.