CivArchive
    Preview 89110
    Preview 98507
    Preview 89666
    Preview 89537
    Preview 98652
    Preview 89667
    Preview 89120
    Preview 89113
    Preview 89118
    Preview 89117
    Preview 89116
    Preview 107335
    Preview 89114
    Preview 107334
    Preview 89112
    Preview 89111
    Preview 89334
    Preview 89333
    Preview 89331
    Preview 98506

    About this model
    Style model for Stable Diffusion. Mainly focusing on graphic design. I am preparing the next iteration - this will be my main focus project. So hopefully we all can benefit soon from next versions.

    Copyrights, Licensing and support
    If you like what I am doing and you want to support my current and upcoming projects: here is my coffee jar

    As I mentioned on my other models - you can use it. Just keep in mind that I spent my money and Time to make it. So if you like it - I will appreciate your support :)
    The only restriction is selling my models.

    Installation:
    As it is model based on 2.1 to make it work you need to use .yaml file with name of a model (vector-art.yaml). The yaml file is included here as well to download. Simply copy paste to the same folder as selected model file. Usually this is the models/Stable-diffusion one.

    Versions:

    There is one version of this model so far. No keyword - you can use it however you like. This is the alpha version.

    Black images issue:

    2.1 models need to have a web-ui config modified - if you are getting black images - go to your config file and add to COMMANDLINE_ARGS= --no-half - potentially it could work with --xformers instead (if supported). This line might slow your generations a bit but will not affect negatively your output.

    Other information:

    Look up prompts on the attached images. Have fun, and experiment with these.

    Description

    FAQ

    Comments (17)

    AiliFeb 11, 2023
    CivitAI

    Great job, looks amazing, can someone please extract a LORA from the checkpoint ?

    SnowwyAIFeb 12, 2023· 2 reactions

    If @pmejna doesn't mind, I'll see about extracting one.

    SnowwyAIFeb 13, 2023

    @Aili I have a LoRA extracting at dim 300 and float16, but I need to get to sleep for work in the morning. Once I get home in the evening tomorrow, I'll do some testing and see how it does. If it works alright, I'll upload it here and post the link. If not, I'll try extracting at a higher dim.

    AiliFeb 13, 2023

    @MrRoboto Thanks man!

    SnowwyAIFeb 13, 2023

    @Aili No problem. I did some experimentation with it just now, and the results are promising. I may still try extracting a higher dim version, though.

    SnowwyAIFeb 13, 2023· 1 reaction

    @Aili Here you are. I went ahead and uploaded the version I extracted last night here.

    MediashockFeb 13, 2023

    @MrRoboto how do you extract LoRA from it? Are there any tutorials out there?

    AiliFeb 13, 2023

    @Mediashock i learned this today actually, here's the tutorial i used however i reccomend using something like 300 dim, not the 128 the tutorial recommends, unless you're extracting from a model that is really over trained i guess

    SnowwyAIFeb 13, 2023

    @Mediashock I used this LoRA extractor by sashaok123 on GitHub. I had to fix a few errors in the Python files, but it wasn't too difficult. I might upload a version to GitHub with the files already fixed when I get home tonight. If I do, I'll link it here.

    Edit: I figure it's easier to just give instructions on how to fix it if you run into the same error I did.

    If you run into the same "ModuleNotFoundError: No module named 'library'" error I was running into, this should fix it:

    1. extract_lora_from_models.py: change line 10 to import model_util

    2. lora.py: change line 11 to import train_util

    3. train_util.py: change line 31 to import model_util

    SnowwyAIFeb 12, 2023· 1 reaction
    CivitAI

    After playing with this some, I think this will come in handy for brainstorming ideas in my own work as a graphic artist.

    pmejna
    Author
    Feb 12, 2023

    thx - I am working now on next version - so far the results are better then current version. Also - both models can create good baseline for mockups. (for can, boxes, business cards). Especially with ultimate SD upscaller. Thx for feedback

    YeiYeiArtFeb 20, 2023· 2 reactions
    CivitAI

    Only works on 2.1 SD? there isnt a version for 1.5?

    THX1139Mar 7, 2023

    I ran it as SD1.5 at 768x768 by mistake the first few times and those were my best results so far. I made different logos in 1.5 and 2.1 tho, so I haven't compared the exact same prompt or seed.

    alifeMar 2, 2023· 6 reactions
    CivitAI

    It always appears after I install

    "Error completing request"

    Please let me know how to fix him

    DarkAgentMar 6, 2023
    CivitAI

    any chance you'd make a 1.5 version?

    pmejna
    Author
    Mar 6, 2023· 2 reactions

    Unfortunately the repo I been using for training SD 1.5 giving me really bad results :/ I will try to find a good one so hopefully in a few weeks I will make one successful

    DarkAgentMar 6, 2023

    @pmejna would love to see it. <3 keep up the good work

    Checkpoint
    SD 2.1 768

    Details

    Downloads
    1,922
    Platform
    CivitAI
    Platform Status
    Available
    Created
    2/10/2023
    Updated
    5/15/2026
    Deleted
    -

    Files

    graphicArt_graphicArtBeta.safetensors

    graphicArt_graphicArtBeta.ckpt

    Mirrors

    graphicArt_graphicArtBeta.yaml

    Mirrors

    CivitAI (83 mirrors)