CivArchive

    LTX 2.3

    Ltx is quite a diva to train... This is my internal 12th revision of the lora. But this one i like. I have decided to take a revision that tends to do more slow motion but with sharp faces and good bounce.

    Enjoy and let me know in comments how you like it, for more speed you can add the word fastly

    Recommended Strength 0.75

    Example Workflow:
    https://civarchive.com/models/1598638/ltx-wan-i2v-22-21-ai-build-art-workflows?modelVersionId=3002323

    Wan 2.2

    V1: Initial Version fast movement less bounce

    V2: More bounce, a little less speed, more WAN 2.1 like

    Works good with my sample workflow:
    https://civarchive.com/models/1598638?modelVersionId=2212578


    This lora is for bounce walk videos. It's quite good in the continuation of the original image.

    Used I2V-14B-720P as base.

    Since ppl were asking about my training Workflow:

    I have used 7 videos with about 4 second length, with faces cut off. I have described what happens e.g.

    Here some captions:

    • A brunette girl, is sitting on the bed of her bedroom and standing up, she wears white bikini, her breasts are bouncing, her face is cut off, she is walking into the direction of the moving camera, noisy image quality

    • A brunette girl, is walking through the floor of her office, she wears a white top and and a black skirt, her breasts are bouncing, her face is cut off, she is walking into the direction of the turning camera


    Then I used Musubi-Tuner with default parameters and trained about 32 epochs and then 16 epochs with larger training set, with my 4090 it was done in about 1h 35 min.

    That's all :)

    Description

    FAQ

    Comments (13)

    andresbravo2003Jun 3, 2026· 1 reaction
    CivitAI

    Fantastic!!

    RedditUser981Jun 3, 2026· 13 reactions
    CivitAI

    sorry for spaming guys bro can i request you to make armpits licking lora for wan 2.2 there is whole community waiting for this type of loras,be the first one to make it

    RobertBobertsonJun 4, 2026

    Some creators charge a very reasonable amount (like $50) to train custom LoRas via their Patreon or otherwise.

    TheEventHorizonJun 4, 2026
    CivitAI

    file says .comfy, not a safetensor? id like someone to verify.

    ai_build_art
    Author
    Jun 4, 2026

    its a safetensor file no worries!

    kenjichan78Jun 7, 2026· 1 reaction
    CivitAI

    great

    darkroast175696Jun 9, 2026
    CivitAI

    I've been having a hell of a time finding good settings for training motion in LTX. If you're willing to share, what settings did you end up with? Learning rate, constant LR or one of the progressive schedulers, DIM and alpha, etc?
    LTX training is so slow it's really frustrating to try over and over.

    ai_build_art
    Author
    Jun 9, 2026

    i was more or less using the default settings, what i have optimized during my iterations was captions and training material, seems like ltx loves leaning the things you dont want it to learn... So for me it was all about the training material,

    darkroast175696Jun 9, 2026

    @ai_build_art Thanks for the reply. Default settings from where? The civitai trainer, or something else? And what lessons did you learn about captions?

    poisenberyJun 11, 2026
    CivitAI

    I sort of figured out the style drift issue with anime images:
    It seems to stem from the low noise model, so disabling that and adding an anime LORA seems to have fixed it for me.
    Hope it helps!

    naomixloveJun 15, 2026

    so just run high noise wan 2.2 model only?

    poisenberyJun 15, 2026

    @naomixlove that's what I did

    VinotintaJul 8, 2026
    CivitAI

    is there a lora like this for the gyatt?

    LORA
    LTXV 2.3

    Details

    Downloads
    3,127
    Platform
    CivitAI
    Platform Status
    Available
    Created
    6/2/2026
    Updated
    7/22/2026
    Deleted
    -
    Trigger Words:
    walking into the direction of the moving camera
    walking into the direction of the static camera
    her breasts are bouncing

    Files