CivArchive
    LTX 2.3 - Multi Step Video Reasoning lora VBVR - v0.1 I2V
    NSFW

    EXPERIMENTAL

    Commercial use is now allowed! Just please credit me :3

    The workflow I use is https://huggingface.co/RuneXX/LTX-2.3-Workflows/blob/main/LTX-2.3_-_I2V_T2V_Basic_for_checkpoint_models.json for some reason the metadata is stripped from the vids idk why. For second and third video I used the res_3s/res_2s sampler, 10eros lora https://huggingface.co/maximsobolev275/LTX-10Eros-LoRA-r768/tree/main and 22b dev model.

    This is a multi step reasoning lora trained on a subset of of the VBVR-MultiStep dataset.

    V0.1 is trained on a 250 video subset from the Physics portion of the dataset. This lora was trained on my LTX 2.3 reasoning merge model that I made by merging my reasoning lora into. This lora stacks well with my other reasoning lora and physics lora and you should notice that it follows prompts better and object to object interactions are better as well.

    Currently working on a version trained on several thousand videos. Potentially larger at a later date. 25k vid version may or may not happen depending on how everything goes. Caching latents on 25k videos takes a enourmous amount of time let alone training.

    I highly recommend using the Sulfur 2 Base model or its rank 768 lora as it makes NSFW generations a ton better.

    Works the same on realistic and non NSFW generations

    Disclaimer & Terms of Use

    This model is provided "AS IS", without warranty of any kind, express or implied, including but not limited to the warranties of merchantability, fitness for a particular purpose, and noninfringement.

    IN NO EVENT SHALL MisticRain69 BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH USE OF THIS MODEL OR ANY OUTPUT GENERATED WITH IT.

    18+ only. You, the user, are solely responsible for anything you generate with this model and for ensuring your use complies with all applicable laws and the CivitAI Terms of Service.

    You may not:

    produce anything illegal, non-consensual, defamatory, or used for harassment, deception, or fraud;

    create any content depicting, or appearing to depict, minors (real or fictional) in a sexual context — no exceptions, ever;

    use this model to depict any real, identifiable person — celebrity, public figure, or private individual — in any context, SFW or NSFW. No likeness use, period.

    If you post outputs to the gallery that violate the CivitAI TOS, you'll be blocked and reported. Don't do this shit. Don't be fucking sus.

    Description

    v0.1 trained on 250 videos

    FAQ

    Comments (8)

    OzzyOsmanMay 9, 2026· 1 reaction
    CivitAI

    Great work!

    HoneyphoriaMay 9, 2026
    CivitAI

    What is better for Eros i2v this or your previous reasoning LORA?

    MisticRain69
    Author
    May 9, 2026

    I would use them both stacked

    jdev1May 9, 2026

    @MisticRain69 thanks. Any recommendations for strength ratio for both?

    MisticRain69
    Author
    May 10, 2026· 1 reaction

    @jdev1 I use them both at 1.0 but u should experiment to see what works best for u

    fluxxesMay 10, 2026· 1 reaction
    CivitAI

    whats multi step reasoning?