CivArchive
    marthedogXkappaspark - V1
    NSFW
    Preview 23799819
    Preview 23799962
    Preview 23800069
    Preview 23820251
    Preview 23820259
    Preview 23820271
    Preview 23820294
    Preview 23820370
    Preview 24909292
    Preview 24909300
    Preview 24909437
    Preview 24909835

    Model Created by combining the Styles of Marthedog and Kappa Spark. Why did I combine these, because from what i've been told 20 pics or whatever is floating around as the minimum to make a LORA is actually not nearly enough and that you want a minimum of 100 if not more to actually get good results. Given this information i combined the more recent works of Kappa Spark and the older works of Marthedog because they have a sorta similar style (IMO, also i really prefer Marthedogs older stuff over his newer style). I'd say the result is decent for my first ever LORA.

    You want to set the weight at around between 0.6 and 0.8 (0.7 gave me the best results) and obviously use a furry Checkpoint like Yiffymix, Furry Rock or Indigo Furry Mix ( The LORA was trained on Furryblend so i'm guessing that'll do best, personally i'm running a merge of Furryblend and Indigomix). Also dunno how or if Pony or SDXL will do a good job with these since was made with SD 1.5 Model but with 1024 set as resolution during training.

    If the artists contact me to do so I will take this model down since I don't want to cause any harm. Just felt like Marthedog in particular is rarely making stuff nowadays and also prefered his older style as mentioned before.

    Description

    Idunno, first time doing this... release i guess?

    LORA
    SD 1.5

    Details

    Downloads
    83
    Platform
    CivitAI
    Platform Status
    Available
    Created
    8/11/2024
    Updated
    4/22/2026
    Deleted
    -

    Files

    marthedogXkappaspark.safetensors

    Mirrors

    Available On (1 platform)

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