CivArchive
    minimax H3 fl2va to ref2va conversion - v1.0
    NSFW

    This loha converts the base text and image to video model to reference to video. It is not perfect, there is difference between result of this loha and what native version gives. The results are not bad however. Just different. (what can you expect with mean cossim of 0.6774)

    loha should be reversable if you set weigth to -1. DO NOT USE JOINT MODEL+CLIP NODE IN COMFYUI. USE THE MODEL ONLY LORA LOADER.

    I used combination of SVD and training.

    Description

    FAQ

    Comments (19)

    ClownimousAug 7, 2026· 2 reactions
    CivitAI

    How does it works ?

    vAnN47Aug 7, 2026· 1 reaction

    from my understanding its takes the ti2va model, add this lora, and it act as ref 2 vid model. question is what workflow because ti2va is only support first last frame, and if you use ref 2 v you can use up to 9 images + video. so my guess is use a ref 2 vid workflow with loading the ti2va model. i'll test it, seems interesting.

    edit: so i tried now an ti2va lora on ref2v and got very bad results, maybe this way you can use an ti2va lora + this lora and a ref2v workflow.. making some generations and will update, maybe even upload some examples if i think about this correctly and get lucky.

    mkDaniel
    Author
    Aug 7, 2026· 1 reaction

    Yes, It converts ti2va to ref2vid. Using this lora you can use ti2va as if it was ref2va. Just a heads up, I did not test the fp8 version

    mkDaniel
    Author
    Aug 7, 2026

    Also in theory it should be able to go in reverse by putting -1 to weigth. Did not test that.

    vAnN47Aug 7, 2026

    @mkDaniel from ref2v model to ti2va behavior? cool

    g1263495582Aug 8, 2026· 1 reaction

    Because both models are the same core model, just fine-tuned for each type of work. If you put the R2V model into an I2V workflow, it will work normally. Most importantly, just write the prompt in the correct H3 format.

    luisa_caoticaAug 8, 2026

    its changes water to h2o

    ShurikAug 7, 2026
    CivitAI

    This seems cool.

    What advantage do we gain using the base model with this lora vs just the ref model alone? I suppose you wouldn't need to load two models if you needed both for a workflow.

    mkDaniel
    Author
    Aug 7, 2026· 1 reaction

    Size, mostly. This is made to free up space on SSD by not needing 2 20GB models.

    From my limited testing (to make sure it works) it can a bit better follow prompt.

    If you only need reference model or only the base text/image to video then this is worthless (might turn reference model to the base one with weigth -1).

    g1263495582Aug 7, 2026
    CivitAI

    I think extracting the LoRA from R2V and using it with FL2VA would probably be better. Honestly, FL2VA can actually do R2V as well, but you have to write the prompt correctly according to the R2V guide.

    g1263495582Aug 7, 2026

    Hold on, why did you use <image2>

    mkDaniel
    Author
    Aug 7, 2026

    @g1263495582 that Is how ref2va points to references

    mkDaniel
    Author
    Aug 7, 2026

    Tried that. Got a mess of pixelů.

    So I Tool ref And the other version, extracted delta Matrix (just ref-fl. Split that using SVD and then again. I continued with training these Split matricies So when combined inside sampler back into delta Matrix And added to fl version it Is as close as possible to the ref version.

    Whole thing took And hour on a 5090. The result Is honestly not that good.

    g1263495582Aug 8, 2026

    @mkDaniel Hold on, what are you mainly using? ComfyUI or something else? And did you write the script you're using yourself?

    g1263495582Aug 8, 2026

    @mkDaniel If it's ComfyUI, use <Picture 1> not <image1>. Also, I suggest checking https://huggingface.co/MiniMaxAI/MiniMax-H3/tree/main/docs first, then rewrite the r2v prompt and re-gen using fl2va.

    mkDaniel
    Author
    Aug 8, 2026

    @g1263495582 Did not notice the comment about the script. Yes, I did. With a help of AI to get better result. This is the first time I am extracting a trained thing when having a base and trained model. I usually train from data, not extract it.

    VesperraAug 11, 2026

    Hey is there any link online for a more detailed explanation on this? I was wondering about how I can use extra references like using a video as an extra reference for the FL2VA model

    calvinAug 8, 2026
    CivitAI

    can you explain why not use the model + clip loader?

    mkDaniel
    Author
    Aug 8, 2026

    It crashes the workflow

    LoCon
    MiniMax H3

    Details

    Downloads
    287
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/10/2026
    Updated
    8/21/2026
    Deleted
    -

    Files

    minimax_h3_lf2ref_full_loha_r256.safetensors

    minimax_h3_lf2ref_full_loha_r256_fp8.safetensors