CivArchive
    malika favre flat color - v1.0
    NSFW
    Preview 1081460
    Preview 1081458
    Preview 1081462
    Preview 1081464
    Preview 1081465
    Preview 1081476
    Preview 1081469
    Preview 1081472
    Preview 1081475
    Preview 1094633

    flat color style illustrator Malika Favre.

    Description

    recommended weight 0.8-1

    FAQ

    Comments (11)

    cd_90Jun 9, 2023
    CivitAI

    That's typical

    XiaoKeai99Jun 10, 2023
    CivitAI

    May I get a lora of fashion clothing illustration? It feels very good-looking. Thank you very much and looking forward to it.

    yuberkley
    Author
    Jun 10, 2023

    can you show some sample pics?

    XiaoKeai99Jun 10, 2023

    @yuberkley OK

    XiaoKeai99Jun 10, 2023

    @yuberkley I sent it to the gallery, sorry I didn't know how to send the pic, so I sent it here

    yuberkley
    Author
    Jun 10, 2023

    @XiaoKeai99 i search with the key word "Anum Tariq", is that style what you want?

    yuberkley
    Author
    Jun 10, 2023

    please check your tailor-made loRA!!

    XiaoKeai99Jun 16, 2023

    @yuberkley thank you very much,its easy to make more works for me

    ploroJul 3, 2023
    CivitAI

    Hey, thanks for putting all those great models out, I am a big fan!

    I've been trying to train a LoRA on flat colour images, a bit like this one and I ran into a few issues. If you don't mind me asking, what base model did you use and how many images did you have for training? I have 172 images that I feel look pretty different from each other but I am still overfitting in less than 2/3 epoch.

    I also noticed that you have a trigger word for this model. Some of the guides I read advise not including any for style LoRAs, any reason you decided to keep one?

    yuberkley
    Author
    Jul 4, 2023

    thank you for liking my loras, obviously, I am not a expert of this area. I don't know what kind of UI you are using for training. I am using the following script to train all my loras.

    https://colab.research.google.com/github/hollowstrawberry/kohya-colab/blob/main/Dataset_Maker.ipynb

    https://colab.research.google.com/github/hollowstrawberry/kohya-colab/blob/main/Lora_Trainer.ipynb

    I used 98 images for the training of this flat colour model, and about 2000 steps and 4 epoch. usually, I use the default setting of other parameters of the script.

    have you tried different weight to see the model overfitting or not?

    talking about trigger word, again, I am not AI expert, I took it for granted all LoRAs need trigger word for my first several LoRA traing. I don't add trigger word during training for my recent ones, I believe it will more convenent for user and myself.

    Good luck and hope to see images generated by your flat colour lora.

    ploroJul 4, 2023

    @yuberkley Thanks :) I will give that a go!

    LORA
    SD 1.5

    Details

    Downloads
    1,943
    Platform
    CivitAI
    Platform Status
    Available
    Created
    6/9/2023
    Updated
    4/30/2026
    Deleted
    -
    Trigger Words:
    favre

    Files

    malikafavre-04.safetensors

    Mirrors

    Huggingface (1 mirrors)
    CivitAI (1 mirrors)

    Available On (1 platform)

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