CivArchive

    I created this LoRA by referring to This.
    I would like to pay my respects to the creator of that LoRA.

    (2026.4.27) Uploaded an LTX 2.3 version LoRA.
    The training techniques previously used with WAN were carried over directly.
    Since this version is a somewhat experimental LoRA, please refer to my sample video in ComfyUI and test it in various ways.
    (This has not been tested in T2V. An NSFW LoRA maybe needed.)
    Again, this is an “experimental” version. Since the step count is low, the results may not be very good. Various tests are currently being conducted, and it may be updated after retraining.

    First of all, I did not train on any facial data. As far as I know, there are already enough (or many) facial-related LoRAs (such as cum on face, cum_in_mouth, etc.).

    If want facial content, please use another great LoRA or use this one together with it (I have not tested the results when using them together).


    -- WAN --
    Using the TripleKsampler node gives even better results. I highly recommend trying it out.
    https://github.com/VraethrDalkr/ComfyUI-TripleKSampler

    Prompt used:
    She moves hand rapidly up and down the shaft of her penis and she ejaculates a thick and huge amount white sticky liquid into air repeatedly from her penis.


    I used 20 videos of 80 frames / 5 seconds each as the dataset.

    Regarding liquid ejection, compared to Wan 2.1 the results are different, and it seems to have become more difficult to generate. However, if you use the workflow of the video I uploaded as it is, the results should not be bad.

    The noise issue in the previous High_noise has been fixed, and older versions have been removed. The noise no longer occurs, and 121-frame sequences are now supported.

    It might also be a good idea to adjust the lightX lora and This lora weights, or change the prompt, to create even better results.

    Description

    ● A somewhat experimental model. The training techniques used for WAN were carried over directly.
    ● Since the training step count is low, it may not work as intended. That is also why I described it as “experimental” Therefore, it maybe updated after retraining.

    FAQ

    Comments (11)

    FantasticFeverApr 28, 2026
    CivitAI

    What if I use this with a hetero couple image. Will the dude turn into a chick?

    razr112Apr 30, 2026· 3 reactions
    CivitAI

    Finally, a good LTX cumshot LoRA! Was the best for Wan, and is now the best for LTX. Awesome work

    DaviDevilJun 20, 2026
    CivitAI

    Hi! Are there any plans to update the LTX model? For example, to help set the precise flow direction. Currently, it's always directed downwards. Liquid can flow where it could logically collide with something along the way. But the image is often completely different when Futa is lying down with her penis near her face—and it's completely impossible to direct the flow toward her face or chest! The stream will still go downwards! Although in this case, it becomes unnatural.

    swan
    Author
    Jun 20, 2026· 1 reaction

    Yes, I’ve received feedback several times since the WAN version about liquid “making contact with” something, so I am aware of that issue.

    For the current LTX version, however, the training data mainly consists of cases where the liquid shoots upward and then falls downward. There is also a small amount of data where it makes contact with the character’s own body, including the chest area. However, it seems that motion involving contact with the character’s own body is not being recognized well.

    The biggest problem is the face. The moment I train the model on data where the liquid reaches the face, it creates a side effect where, regardless of the emission position, the model starts producing semen-breath coming out of the mouth.

    An update would only be possible once this issue is solved, but it is not an easy problem to fix, so I cannot give any definite timeline.

    DaviDevilJun 20, 2026

    @swan I'm very fond of your work! For both Wan and LTX! They're simply indispensable in my work. And mouth spitting is a problem with any other Lora for Wan. But not with yours! And I agree that introducing this to LTX would be a critical issue! LTX has a poor understanding of physical space, so you need to approach this carefully!

    I wish you good luck and look forward to your updates! Whenever they come.

    gokgokgokJun 24, 2026
    CivitAI

    any tips to reduce the amount of cum?

    MorphMuseJun 24, 2026· 1 reaction
    CivitAI

    What settings/prompt are you using for the cum to come out of her penis? Mine seems to be random and no where near it. Is there also a way to time when it happens? That's also random for me.

    paraverpornoentwiterJul 7, 2026· 1 reaction
    CivitAI

    sin duda alguna es el mejor lora para CUm..ts, necesitamos la vercion V2.0 esta genial solo que aveces el sem..en se comporta antinatural a,veces igual gran aporte mis felicitaciones.

    PracticalLetter44912Jul 13, 2026
    CivitAI

    I like this lora very much so. When applying, however, some of the things in the image are shaking in the same rhythm as masturbation. Any idea how to avoid?

    swan
    Author
    Jul 13, 2026

    I haven’t experienced that issue myself, so I’m not sure how helpful this will be, but you could try lowering the LoRA strength a little. Also, try writing the prompt so that only the specific object you want moves. LTX is especially sensitive to prompt wording, so prompts need to be very precise.

    @swan Oh, thank you, good advice