CivArchive
    Text Logo - Style - V1
    Preview 25412426
    Preview 25411791
    Preview 25411789
    Preview 25411612
    Preview 25412489
    Preview 25411788
    Preview 25407944
    Preview 25407946
    Preview 25407945
    Preview 25412479
    Preview 25412699
    Preview 25413592
    Preview 25413598
    Preview 25414371
    Preview 26166260
    Preview 26166368
    Preview 26167081
    Preview 26167113
    Preview 26169695

    Trained on 300 images of various words spelled in different styles, materials, fonts, etc. with the text front and center as the primary focus.


    Try starting your prompt with the word ...

    like this example:

    the word "DAY", made of clouds in the sky, night time with a crescent moon above, cinematic angle and dramatic lighting, text focus


    Top tags:

    text focus, black background, sky, flower, cloud, gradient, still life, dripping, slime, leaf, food, space


    Most common words depicted in training data:

    boy, no, yes, hot, day, hope, trust, cold, girl, dude, more, test, air, home, slow, take, candy, fast, one, how, move


    Most common themes in training data (in no particular order): sky/clouds, simple monochrome font with gradient or black background, flowers, slime, cars, snow, fire, space, autumn leaves, bacon

    Description

    Trained using Civitai

    FAQ

    Comments (3)

    mememoryAug 25, 2024· 1 reaction
    CivitAI

    More consistent fonts and letters alignement, without breaking the original prompt. That's not bad ! Thank you.

    GitarooManSep 14, 2024· 1 reaction
    CivitAI

    Pretty cool! What lora strength and base Flux models do you recommend?

    creedukFeb 26, 2025· 1 reaction
    CivitAI

    Worked great tested several others, (added this comment on wrong one to start with but double checked and it is Text Logo - Style V1 LoRA).

    Tested with fixed seed with FLUX Fill (which may be more challenging?) Made bigger changes with the trigger word , but ironically I had the wrong trigger word and got really nice results "perfect text title style " I need to run this on others to see but was the first 1 from about 20 tests on that seed and mask to be clean, clear, sharp. I then tested same prompt with that trigger and LoRA off just to be double sure it was not the prompt doing the hard work, but no it was the combo with this LoRA, it is working. A lot of FLUX LoRAs do minimal or nothing we have to remember to check they make a difference :)

    The only grumble I had with my output which might be FLUX? Is a scribble like representation of a TM and Copyright symbol added as well. Prefer none and add them later if desired. Like I say that might be from the model. Maybe a way to prompt out?

    LORA
    Flux.1 D

    Details

    Downloads
    1,455
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    4/24/2025
    Updated
    5/1/2026
    Deleted
    4/24/2025
    Trigger Words:
    text focus

    Files

    749547_training_data.zip

    Mirrors

    CivitAI (1 mirrors)

    Available On (1 platform)

    Same model published on other platforms. May have additional downloads or version variants.