CivArchive
    Thick Lips Blowjob - v1.0 Low
    NSFW

    My first attempt at training a lora. It turned out well enough to share, I think. Trained on 20 5-second clips at 512x512 resolution. The intention is for it to be a lora with more of a focus on thick lips working the head of a penis. Trained primarily on pov shots, with some three-quarters and side shots. If you have tips for motion lora training, I'd love to know!

    It's not the most consistent at forming a good-looking glans when there isn't one in the initial image, so I recommend pairing it with the low noise side of a lora good at that, such as Ultimate Deepthroat. It seems to work well enough without the trigger word.

    All showcase videos were made using this lora on high and low at 0.8-1.0 strength, but I recommend testing different strengths. My preferred generation method is to use multiple loras at lower strengths, but I wanted to clearly demonstrate this one in the previews.

    Generations in the showcase use 4 steps high, 4 steps low. They also use the following:

    Workflow: Smooth Workflow v2.0 - https://civarchive.com/models/1847730?modelVersionId=2264611

    Checkpoint: Smoothmix i2v 2.0 - https://civarchive.com/models/1995784?modelVersionId=2513182

    High Noise lightx2v 1022 - https://civarchive.com/models/1585622?modelVersionId=2337890

    Low Noise lightx2v (1.5 strength) - https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Lightx2v/lightx2v_T2V_14B_cfg_step_distill_v2_lora_rank128_bf16.safetensors

    Description

    Initial low version

    FAQ

    Comments (29)

    GlowingGuardianGirlJan 19, 2026· 2 reactions
    CivitAI

    Thank you for sharing. I just started to use Wan and I need some advice, if you ever have time, could you DM me?
    First question, why do you use 1.5weight for the LowNoise Lightx LoRA? thank you 🙌

    Alberist
    Author
    Jan 19, 2026

    Hi there. It's what's recommended by the author of the SmoothMix checkpoints. I recommend playing around with lora weights as much as you can afford to, as it's rarely an exact science.

    GlowingGuardianGirlJan 19, 2026

    @Alberist Ow ok. I'm just trying on a very old 8gb VRAM gpu and I'm getting a lot of black outputs, trying to figure out why by playing with steps and weights but damn it's harder than I thought to get wan to give decent outputs on low vram.
    Does he give an explaination? Is it to compensate for something during the low noise steps? Thank you for your answer

    Alberist
    Author
    Jan 19, 2026· 1 reaction

    @GlowingGuardianGirl Oh, I didn't realize you weren't aware of how lightx2v loras worked. To put it simply, they're kind of like a simplified version of the overall WAN model. By using them, you're able to use fewer steps to generate your video. They help the noise resolve into a final result faster at the cost of a little bit of quality and prompt adherence. For that one in particular, videos come out a little fuzzy if you don't use enough strength. I believe it needs 1.5 strength because it was made for Wan 2.1. Newer lightx2v loras usually need 1.0 strength, but I find that one works well with Smoothmix 2.0, so I use it.

    If you're getting black outputs, you may need to update comfyui. I had that issue a while ago, and it turned out something I was using required a newer version of comfy.

    GlowingGuardianGirlJan 19, 2026

    @crafted101 Might actually try that. GPT can be very stupid. [damn you changed your nickname!]

    Alberist
    Author
    Jan 19, 2026

    @GlowingGuardianGirl Yeah. Grok or Claude are good free options for that. Both can do web searches and are usually pretty helpful in troubleshooting issues.

    NyxxiNyxJan 19, 2026· 7 reactions
    CivitAI

    Oooooh. Looks good.

    Downloading to test. We can never have enough good blowie LoRAs. ;)

    sirbazzrickJan 19, 2026· 2 reactions
    CivitAI

    excellent work! and it doesn't change the face which is great... looking forward to your future work

    StrangeLiminaJan 19, 2026· 3 reactions
    CivitAI

    This is exceptional, good work!

    210881175Jan 19, 2026· 2 reactions
    CivitAI

    DAMN good!

    fatberg_slimJan 19, 2026· 7 reactions
    CivitAI

    Great job on your first LoRA. Congrats! I did some testing with a GGUF base model and you absolutely need another LoRA for the glans when the penis is already in her mouth. else it'll look like a sausage. I chose Oral Insertion 0.5 on H and L and with this combo I'm blown away how great this LoRA is. It works fine when the glans is visible in the base image though. But since I'm lazy I kept the LoRA-Combo for all my runs.

    Thanks so much, Alberist. 🤜🏻🤛🏻

    Another fine addition to my evergrowing BJ/Deepthroat LoRA-Stack.

    StrangeLiminaJan 19, 2026

    It's interesting that you say this because in my experience "already in her mouth" just basically never looks very good with any combination of LoRAs I try, at least nowhere near as good as if you start with the full context in the base image.

    Alberist
    Author
    Jan 19, 2026· 1 reaction

    @StrangeLimina I find the are a few loras that produce good looking glanses if you use their low noise lora on higher strengths. Though you definitely get more detail and control if you just have it in the initial image.

    fatberg_slimJan 19, 2026· 1 reaction

    @StrangeLimina For me Oral Insertion makes the glans look good most of the time.

    StrangeLiminaJan 19, 2026· 1 reaction

    @fatberg_slim @Alberist Interesting I'll give that a try, thanks for the tip (heh).

    TwoMoreLurkerJan 19, 2026· 4 reactions
    CivitAI

    Very well done, especially for a first Lora!!

    boobkake22Jan 19, 2026· 2 reactions
    CivitAI

    Nice job identifying a need. Most LoRA's don't seem to have enough face diversity to capture this. Looking forward to trying this one.

    MGHerderJan 19, 2026· 1 reaction
    CivitAI

    I'm testing. So far all my gens have been very different than everyone else -- tons of head bobbing, like 6 bobs in a 7 second clip. Been tweaking strength and it seems to need a little help LoRA for the glans as Fatberg was saying. (I went with experimental NSFW).

    Alberist
    Author
    Jan 20, 2026· 1 reaction

    Yeah. Oral insertion, Ultimate Deepthroat, and Big Cock Oral should all be reasonable ways to improve penis shape, depending on your preference.

    MGHerderJan 20, 2026· 1 reaction

    @Alberist Your LoRA takes "she is bobbing her head multiple times" and runs with it! Maybe I need to adjust the shift to get it to work. But taking that out, and I'm getting the real thing. Again, great job on this LoRA.

    cornfloursandboxJan 20, 2026· 2 reactions
    CivitAI

    Top notch bj lora this, 10/10. Good cohesiveness, minimal facial morphing. Will certainly be using it lots in future, combined with dreamley and the penis glans was just fine for me.

    MrHappyHappyFunFunTimeJun 3, 2026

    You could train one and make it Cornhub

    TheLuckyBunnyJan 21, 2026· 2 reactions
    CivitAI

    Hey there^^, can i ask what setup you used to train this model? (GPU/RAM/CPU).

    Alberist
    Author
    Jan 21, 2026· 1 reaction

    I rented an H200 for a little over a day and trained in the cloud.

    TheLuckyBunnyJan 21, 2026

    @Alberist ah ok, i was trying to figure out if i could somehow do it with a 5090 but until now no good resaults sadly

    Alberist
    Author
    Jan 21, 2026· 3 reactions

    @TheLuckyBunny I wouldn't give up! @TwoMoreLurker put out this guide on lora training on a 3090, and their results still come out good. https://civitai.com/articles/18985/wan-22-local-lora-training-guide-windowslinux

    I do plan on trying to train with my own 5090 in the future, but I'm usually running generations at any given time.

    TheLuckyBunnyJan 22, 2026

    @Alberist oh thats awesome, thanks! will look into that the next few days. if you are ever interested in doing a cooperation (maybe trying to develop new character or animation models or working on workflows) i am always open for it, currently i am mainly working on a Discord where i share stuff that i make and where people can request stuff.

    quadrapoleMay 3, 2026

    i have 2 rtx 6000 pro. is it possible to train with this hardware?

    Alberist
    Author
    May 3, 2026

    @quadrapole Yes, should be completely possible. I believe you could do it with as little as a single 3090 if you're patient enough.

    LORA
    Wan Video 2.2 I2V-A14B
    by Alberist
    Download (Beta) View on CivitAI
    oral
    action
    lips

    Details

    Downloads
    5,328
    Platform
    CivitAI
    Platform Status
    Available
    Created
    1/18/2026
    Updated
    8/3/2026
    Deleted
    -
    Trigger Words:
    lips-bj

    Files

    lips-bj_low_noise.safetensors

    Size:
    146.34 MB
    SHA256:
    2fb4bc0fe0d47c87384ee8c4376602547981127a0da5b01500557a92405ed718

    Mirrors

    HuggingFace (53 mirrors)
    CivitAI (1 mirrors)