CivArchive
    KevaMixFlat - v1.0
    Preview 2525110Preview 2525109Preview 2525084Preview 2525094Preview 2525101Preview 2525106Preview 2525098Preview 2525097Preview 2525102Preview 2525107Preview 2525108Preview 2525105Preview 2558448Preview 2573982Preview 2575290Preview 2589075Preview 2589661Preview 2590155Preview 2603244Preview 2603252

    INTRODUCTION

    I created this model just for fun. I had been working on it for a while and had been enjoying generating images with it. I wasn't sure if I should share it because I'm certain that other models can produce similar results, but I decided to share it anyway because I'm having fun with it.

    What did I use to make this mix ?

    I used a variety of existing anime models, and while I don't remember all of them, I do recall that the last one I used was Flat-2D Animerge

    How I use this model ?

    Sampler: For all my images, I use DPM++ 2M Karras, you can use whichever one you prefer; I haven't noticed any major differences between them.

    Steps: I typically use a range of 20 to 30 steps, but you can experiment with higher values if you think they might yield better results.

    CFG Scale: I personally don't change it.

    VAE : I use these two "kl-f8-anime2.ckpt" and "vae-ft-ema-560000-ema-pruned.ckpt", but feel free to use any VAE of your preference.

    Upscaler: I use the 4x-Ultrasharp but you could also use R-ESRGAN 4x+ Anime6B, Personally, I prefer 4x-Ultrasharp

    Description

    Checkpoint
    SD 1.5

    Details

    Downloads
    767
    Platform
    CivitAI
    Platform Status
    Deleted
    Created
    4/24/2025
    Updated
    5/6/2025
    Deleted
    4/24/2025

    Files

    kevamixflat_v10.safetensors

    Mirrors

    Huggingface (1 mirrors)
    CivitAI (1 mirrors)