CivArchive
    THM_UNC_Z-image_TE - v1.0
    NSFW
    Preview 131814932
    Preview 131815045
    Preview 131815258
    Preview 131815404
    Preview 131815403
    Preview 131815410

    This is an uncensored fine tuned text encoder for Z-Image base and Z-image Turbo.

    Z-Image Turbo model used: https://civarchive.com/models/2644869/solordz?modelVersionId=2969766

    Z-Image base model used: (To Be Released) https://civarchive.com/models/2655305/lord-zolaris Watch this space

    Workflow:

    https://civarchive.com/models/2662358/lord-zolaris-no-upscale-needed-workflow

    Also check out https://civarchive.com/models/2662833/zimagetesolordz?modelVersionId=2990083

    We work together and this is her version at the moment!

    We decided to release the latest current version of the text encoder for free, despite the amount of time, testing and effort that went into it. If you would like to support us for future releases (of which there will many), for now you can donate to me directly here:

    https://www.buymeacoffee.com/tripleheadg

    We should have a Patreon or something equivalent set up in future for early access to our new model files directly. Don't bother sending me Buzz because I am located in a country which cannot recieve monetary rewards from Civitai.

    We used our own custom training code and custom nodes for merging.

    Z-image Base^

    Z-image Base^

    Z-image Turbo^

    Z-Image Base^

    Z-Image base^

    Z-Image Base^

    Z-Image Turbo^

    Z-Image Base^

    Z-Image Base^

    Z-Image Turbo^

    Description

    Works for both models.

    FAQ

    Comments (8)

    wickedywakMay 26, 2026· 5 reactions
    CivitAI

    I was skeptical that this would make much of a difference at all, but am pleasantly surprised how well it works. Thanks!

    Triple_Headed_Monkey
    Author
    May 26, 2026

    Awesome! I'm glad you like it!

    WullyBully69Jun 4, 2026· 7 reactions
    CivitAI

    So i've tried at least 8 different TEs with varying results, but after much experimentation and testing i'd have to say that yours works extremely well for me. In my setup, I use LM Studio to auto generate prompts based on image inputs, and your TE takes those prompts and reproduces the images (based on the models I use) more accurately than any of the others I have tried. Thanks for all your work on this.

    Triple_Headed_Monkey
    Author
    Jun 5, 2026

    I’m really glad that it works well for you. :) Thanks for testing and sharing your experience :)

    thinotg0748Jun 10, 2026

    How do you use LM studio to auto generate prompts? Please do you have any tutorial

    WullyBully69Jun 10, 2026· 4 reactions

    @thinotg0748 I was considering making one but I haven't yet. There is a NODE pack for comfy that I use that links to LM studio. Then you just have to download the right VL models inside LM Studio. If I do make a tutorial with my workflow, I will post it here on Civ.

    thinotg0748Jun 12, 2026

    @WullyBully69 Thanks for the tip and definitely looking forward to it

    northruneAug 21, 2026

    you can wire LM studio into your workflow with: https://github.com/EnragedAntelope/EA_LMStudio

    TextEncoder
    ZImageTurbo

    Details

    Downloads
    2,314
    Platform
    CivitAI
    Platform Status
    Available
    Created
    5/25/2026
    Updated
    8/30/2026
    Deleted
    -