CivArchive
    Preview 2604631
    Preview 2604651
    Preview 2604657
    Preview 2604658

    Welcome to Rae-Series!!

    RaesanMix is a merged model that focused in 3D anime model with more flexibility


    Enjoying my work? Consider supporting the development of my models by buying me a coffee on Ko-fi

    Also check out my other models on Mage: Raepix Focus XL / Raepix Focus XL Lightning / Raevitamix Realistic / Raevista XL


    Suggested Prompt

    use this for starting point

    (masterpiece, best quality),

    Negative

    (worst quality:1.4, low quality:1.4, normal quality:1.4), lowres, blurry, monochrome, zombie, (interlocked fingers:1.2),
    • Sampler : Euler a or any DPM++

    • Steps : 20+

    • CFG Scale : 7+

    • Clip Skip : 2

    • Vae : included, set it to auto

    • Resolution : 512x904 (this is just my preference), Standard 512x768

    Hires.fix Settings

    • Upscaler : 4x-AnimeSharp

    • Hires step : 10~15

    • Denoising : 0.25~0.45

    • Hires scale : 2

    Optional

    • Set eta to 0.667

    • Extension !ADetailer can help to fix face if its too far

    Resource used to merge :

    RaenaMix + RaelisMix

    RealCartoon 3D by 7whitefire7

    DucHaitenNiji by DucHaiten

    RichyrichMix by richyrich515

    DarkShushi2.5D by Aitasai

    MergeBlockWeighted

    Description

    Improved body feature

    FAQ

    Comments (10)

    265275Oct 9, 2023· 2 reactions
    CivitAI

    Tested all your models. They are all great, congrats! I can finally move on from YesMix1.6 xD

    Raelina
    Author
    Oct 10, 2023

    Nice, thank you for testing out all of my models.

    265275Oct 13, 2023

    @Raelina Only thing I noticed, and I don't know if it's a skill issue of mine, but I find it hard to get expressions out of this model. The other ones seems to be easier.

    Raelina
    Author
    Oct 13, 2023

    @FroggyWoggy Can you tell me more specifically which expression and what prompts you use? So I can try and test it. If the problem is within my model, I will try to fix it in the next update.

    265275Oct 17, 2023

    @Raelina using this prompt as a example: "(masterpiece, best quality), intricate details, underwear, face close up, indoors, night, mature female, breasts, disgust:1.4, looking at viewer". No matter how much weight I add, it does not reflect in the images. This also happen to "disdain" and "serious". These are the only expressions I've seen this happen. Feel free to tell me if I'm just being stupid and missing something. Thanks for your hard work with these checkpoints

    Raelina
    Author
    Oct 17, 2023

    @FroggyWoggy look at my newest post gallery,, is that expression you mean.?

    265275Oct 17, 2023

    @Raelina indeed. I guess I did not put that much weight into it? I went up to 1.3 in my testing. Is this because datasets have way less images with this expression? I'm not very familiar with finetuning of checkpoints

    Raelina
    Author
    Oct 17, 2023· 2 reactions

    @FroggyWoggy when you need to put more weight, you need these "()". without these AI will recognize it just normal tags even you already put 1.6 or so. ex: (serious), (disgust), (serious:1.2), (disgust:1.4) etc

    265275Oct 17, 2023· 1 reaction

    @Raelina well, now I feel kinda stupid lmao. I never knew that using :1.6 without the "( )" would be meaningless. Thanks for the lesson, and sorry for giving you the trouble of investigating this.

    Raelina
    Author
    Oct 17, 2023· 1 reaction

    @FroggyWoggy no worries

    Checkpoint
    SD 1.5

    Details

    Downloads
    1,678
    Platform
    CivitAI
    Platform Status
    Available
    Created
    9/21/2023
    Updated
    5/12/2026
    Deleted
    -

    Files

    raesanmix_v20.safetensors

    Available On (1 platform)

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