CivArchive
    Assisted handjob [Real] | Wan 2.2 I2V - v1.0 [high]
    NSFW

    Hey babe, stroke it 😏

    Of course darling, on my way 😊


    Wanna support me?

    Don't send Buzz, send me USDc or USDT through Polygon Network:

    0x3f1e1323fa90b4c9f5a386bfa5b1e72f5a6c25e9

    $0.10 or $10, any amount is welcome and helps me make more LoRAs for the community


    Workflow

    I don't use ComfyUI, so I don't have a workflow to provide πŸ€·πŸ»β€β™€οΈ

    The training is done on Wan 2.2 A14B Diffusers (BF16)


    Training

    • 480 steps on high noise

    • 3 edited videos (80 frames each video)

    • HLR + LCD (These acronyms were created solely to confuse you)

    • Lots of hope

    • $8 (βœ‹πŸ»πŸ˜­πŸ€šπŸ»)


    Usage (Low noise)

    This LoRA was trained only on High Noise, which means you'll have to use some Low Noise that knows what a penis is (I used the Low Noise from DR34ML4Y V2).


    Got an idea for LoRA? Then let's do it! (or at least try 🀠)

    In my repository, I created a model called "LoRA Requests", so if you want to request a LoRA, send videos and photos to help build the dataset for a public LoRA that will be posted here, that's the place to go.

    Open LoRA Requests on HuggingFace πŸ€—


    I2V:

    Prompt (ongoing):

    Then she holds his hard penis, she moves her hand up and down with her fist closed, stroking repeatedly.

    Prompt (cut to revealing hard penis):

    No man, at least not here 🀫

    Prompt (cut to stroking hard penis):

    No man, at least not here 🀫

    High Noise LoRA Scale: 1.0

    Low Noise LoRA Scale: 1.0

    Shift: 4

    To see all uncensored prompts (cut to action), visit this LoRA on HuggingFace πŸ€—


    T2V:

    Technically it works, but I haven't tested it. If you're going to test it, use a low scale, starting at 0.5 on High Noise LoRA.

    Prompt (caressing bulge, maybe not working):

    Medium shot of her and a man sitting on a gray sofa in a living room with white walls. The man has fair skin, messy hair, a beard on his chin, and a mustache. He is wearing a black shirt with a white design and gray shorts with a hard bulge. He is relaxed on the sofa with his legs open, one hand resting on her shoulder, caressing it. She has fair skin, long black hair, a tattoo on her wrist, and a ring on her left ring finger, wearing only a very short white mini top.
    She is lying on his chest, holding and stroking his hard bulge, moving her hand up and down. Her expression is neutral, avoiding looking at the camera.
    

    Prompt (revealing penis):

    Medium shot of her and a man sitting on a gray sofa in a living room with white walls. The man has fair skin, messy hair, a beard on his chin, and a mustache. He is wearing a black shirt with a white design and gray shorts with a hard bulge. He is relaxed on the sofa with his legs open, one hand resting on her shoulder, caressing it. She has fair skin, long black hair, a tattoo on her wrist, and a ring on her left ring finger, wearing only a very short white mini top.
    She's lying on his chest, holding his hard bulge, then she pulls up his shorts, revealing the hard penis underneath. Her expression is one of surprise.

    Prompt (stroking hard penis):

    Medium shot of her and a man sitting on a gray sofa in a living room with white walls. The man has fair skin, messy hair, a beard on his chin, and a mustache. He is wearing a black shirt with a white design and gray shorts pulled down with a hard penis. He is relaxed on the sofa with his legs open, one hand resting on her shoulder, caressing it. She has fair skin, long black hair, a tattoo on her wrist, and a ring on her left ring finger, wearing only a very short white mini top.
    She's lying on his chest, holding his hard penis, she moves her hand up and down with her fist closed, stroking repeatedly. Her expression is one of concentration.

    You can change or add details, but do not change the prompt structure.


    About HLR + LCD

    This is a fast-learning technique, which makes LoRA less flexible. This can drastically reduce creativity, but it yields stable results while using few resources. Negative effects:

    1. It will probably only work at the same camera angle

    2. High chance of not responding to different prompts

    3. High chance of forcing the original characters of the training video Training only the High Noise reduces this chance, but it can force things that are VERY similar, like the same hair length, same color, etc.


    Download

    Download High Noise LoRA (On HuggingFace)

    Download Low Noise LoRA (DR34ML4Y_I2V_14B_LOW_V2)


    Help me creating more

    If you want to help me continue making LoRAs, or if you want me to make a LoRA for you, buy 5000 PlayCoins at Aisha-AI and transfer them to my account (account number 2).

    This helps Aisha-AI to stay alive and produce new LoRAs for you all πŸ’œ

    Description

    First version

    FAQ

    Comments (12)

    Glaciousg1acx23Mar 7, 2026Β· 4 reactions
    CivitAI

    Awesome work!!! Just one little thing, I'm using the same Lora you mentioned in the low version and yours in the high version but the faces do keep changing quite a lot. is there anything I can do to help with your Lora not changing the faces as much?

    AishaAI
    Author
    Mar 7, 2026Β· 3 reactions

    You're probably referring to "cut to," in which case try adding You're probably referring to the "cut to" sequence; in that case, try adding elements that remind viewers of the woman's face from the initial image, or indicate that she is the focus.

    "The scene begins with a woman posing, she is the focus of the scene."

    Another possibility is to describe the details of it.

    Another possibility is to describe its details at the beginning, but if I posted the LoRA without a warning, it's because it's working normally.

    In this case, you may not be using the standard Wan 2.2 workflow.In that case, you might not be using the standard WAN 2.2 workflow. I train all my LoRAs on WAN 2.2 A14B Diffusers BF16, without lightning and without changing CFG or steps.

    fenasikerimMar 7, 2026Β· 7 reactions
    CivitAI

    Is it just me or does "cut to" have 0 face consistency?

    AishaAI
    Author
    Mar 7, 2026Β· 2 reactions

    Are you seeing different people posting results with facial consistency?Are you seeing different people posting results with facial consistency?

    I am

    AI4L1VEMar 7, 2026Β· 1 reaction

    It's not just you. It simply does not work. He just claims it does and tells you, you are doing something wrong. Yet there is not a single example of other people where it actually works.

    AI4L1VEMar 7, 2026Β· 5 reactions
    CivitAI

    Please stop claiming the "blink" i2v works. It doesn't. In my Discord channel, a dozen people tried it with different settings and wf's and prompts. It simply does not work, stop pretending please.

    AishaAI
    Author
    Mar 7, 2026Β· 2 reactions

    I just recorded my screen doing a test, generated 2 videos and both videos came out perfect.

    I'm just waiting for Civitai to catalog and make it public, I'll send the link here soon πŸ˜‡

    AishaAI
    Author
    Mar 7, 2026Β· 3 reactions
    tipholditdown513Mar 8, 2026
    CivitAI

    again, amazing work. (really thinking about requesting a custom lora lol) but could you share the workflow you used to generate such videos? I'm always getting different results that doesn't look like the examples you've created

    AishaAI
    Author
    Mar 8, 2026Β· 2 reactions

    I use the Replicate API, and they use the standard WAN 2.2 workflow; at most, there's FP8 quantization.

    Dumcluck51Mar 9, 2026
    CivitAI

    SO many penis loras have the glans on backwards. I tried this one and it also has that problem. Maybe not important to most but to me it is an instant ugghhh.

    AishaAI
    Author
    Mar 9, 2026

    Open XVIDEOS and be happy staring at real glans 😏