CivArchive
    Super Realistic POV Oral Sex for Hunyuan Video - v2.0
    NSFW

    Man! What can I say?

    This is absolutely fantastic! You just have to try it to truly see the difference compared to the others.

    V2 has released without face locking and more different action on Oral performance!

    There is workflow metadata in the videos! Right click and choose "save video as", and NOT "save videoframe as" which is indeed empty


    Support me:

    # Support Me!

    If you think it's good, remember to support me!

    Here are my contact details:

    - Civitai: [ttplanet](https://civarchive.com/user/ttplanet)

    - Discord: ttplanet

    - X: @ttplanet

    Contact me if you want your customized lora!

    Highlighted Features

    • Full-Speed Movement: No sluggish actions like most LoRA or video models.

    • Realistic Style: Delivers a genuine POV experience.

    • Customizable Characters and Backgrounds: Easily replace or combine them as you like.

    • Super Realistic Expressions: Natural-looking faces with eyes properly focused on the camera.

    • High Robustness: Almost a 100% success rate in generating videos!

    • it also works for man characters....


    Areas for Improvement

    • Facial Variability: Currently, faces are too rigid, especially for Asian features. Will address this in the next update.

    • Proportionality: Some generated elements appear oversized (e.g., certain body parts). Planning to retrain with more appropriately scaled samples. If you can provide relevant material, it would be greatly appreciated!

    Description

    Thanks for the support. V2 version is free released

    FAQ

    Comments (13)

    LatentDreamJan 7, 2025· 1 reaction
    CivitAI

    thanks for pointing us to those lora loaders.
    i havent really understood how to use that lora loader in "double blocks" mode or "all" mode, can you expand more on that so i can add this infos on my article? https://civitai.com/articles/9584 whats the difference between double and all?

    ttplanet
    Author
    Jan 7, 2025

    pelase download my cover as sample, you will understand the workflow

    LatentDreamJan 7, 2025

    @ttplanet the workflow itself do not explain how that function works but never mind..
    i discovered it by myslef after like 1 hundreds tests comparisons.
    it turnes out double blocks do not always be the best option, i don't know why everyone saying it is.
    thanks aniway

    georgelukJan 7, 2025

    Is there a way to do it using the default comfyui options? I'm a bit apprehensive about downloading a node with so little info. Especially after that one node from months ago that turned out to have a virus.

    ttplanet
    Author
    Jan 7, 2025

    @LatentDream I will Say all for character double for motion

    LatentDreamJan 8, 2025

    @ttplanet yes this is kinda my conclusion too but still don't know exactly what means

    bbeggs234Jan 8, 2025· 2 reactions

    @georgeluk was it from github? And what node you are talking about?

    WhatTheGuyJan 8, 2025· 1 reaction
    CivitAI

    V2 is a very nice update. I'm getting a lot of new actions that weren't there before in V1 =)

    ttplanet
    Author
    Jan 8, 2025

    thanks for the buzz support in the beginning,I have to make it better

    6010749Jan 8, 2025· 2 reactions
    CivitAI

    Should train using Sladkislivki as a reference.

    AiMetatronJan 11, 2025· 3 reactions
    CivitAI

    上帝已经不忍直视了 >_<

    ziriuss1Jan 12, 2025
    CivitAI

    Hello again, checking the workflow I see that you use a diffuser model called p_rank_00_model_states.pt, where I download it because those who use all have the name Hunyuan-Video-T2V-720p, and also use GGGF but is not compatible with this workflow

    ttplanet
    Author
    Jan 14, 2025

    Official bf16 weight, I just did not change the name