CivArchive
    ThisIsCartoon - v2.0
    NSFW
    Preview 3072889
    Preview 3072892

    Support (Patreon)

    Pruned Version 3 is now available

    VAE: VAE-FT-MSE-840000-EMA

    ThisisCartoon is a mixture of anime and hentai

    Version 3

    • Better representation of the character (cartoon-like)

    • Better representation of the faces

    • Better representation of different places/objects

    Version 2

    To create my 2 versions, I took my First version as a base framework and trained those two with multiple images, so MERGE.

    I trained the 2 versions again with about 40,000 to 45,000 images.

    With a size of 768x768 and 1024x1024.

    Version 1

    For the creation was used an anime/hentai database with about 75,000 images in different positions, character, places and creatures.

    The size was trained with 768x768 and 1024x1024

    Description

    Better representation of the character (cartoon-like)

    Better representation of the faces

    Better representation of the hands

    Better representation of different places/objects

    FAQ

    Comments (7)

    TiranaOct 21, 2023· 1 reaction
    CivitAI

    What is the difference between V2 and V1 version?

    ChangeMeNot
    Author
    Oct 22, 2023

    Better representation of the character (cartoon-like)

    Better representation of the faces

    Better representation of the hands

    Better representation of different places/objects

    TiranaOct 22, 2023· 1 reaction

    @ChangeMeNot OH! Thanks you for your information.

    settima_aiOct 21, 2023
    CivitAI

    Interesting story. I dont understand a point.

    It's a "mix" but you also said "anime/hentai database with about 75,000 images" was used.
    So you made a training in addition to the mix?

    ps: where is the version 2 pruned? I dont see it!

    hollamanOct 21, 2023· 1 reaction

    I am also curious about this. You can find both pruned and full versions on the right hand side under "Files" though :)

    ChangeMeNot
    Author
    Oct 22, 2023

    To create my 2 versions, I took my First version as a base framework and trained those two with multiple images, so MERGE.

    I trained the 2 versions again with about 40,000 to 45,000 images.

    With a size of 768x768 and 1024x1024.

    What was important to me with the 2 versions

    1. that the images look more like a cartoon.

    2. the representation of the faces

    3. representation of hands

    4. representation of different places/objects

    hollamanOct 22, 2023· 1 reaction

    @ChangeMeNot Thanks for the explanation. So merge, but finetuned too. The results are great!

    Checkpoint
    SD 1.5

    Details

    Downloads
    413
    Platform
    CivitAI
    Platform Status
    Available
    Created
    10/21/2023
    Updated
    6/20/2026
    Deleted
    -

    Available On (1 platform)

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