all / g / s / fur / art / ai / 3dcg / f / rp / req / p2p / dis / lit / kaki chat
[Settings] [Home] [Catalog] [Search] [Thread list] [Stats] [Reports] [Watcher] [PMs] [Admin]
[Return] [Bottom]

Posting mode: Reply
BBCode
(for deletion)
  • Allowed file types are: gif, jpg, jpeg, png, webp, webm, mp4
  • Maximum file size allowed is 25000 KB.
  • Images greater than 255 * 255 pixels will be thumbnailed.
  • 107 unique users in the last 30 minutes (including lurkers)
    First
    [1]
    		Last
    File: 00002-3989541387.png
    (1.84 MB, 1200x1200) ▶
    1934098
    No.13921 [Report] AS
    ▶
    OP from >>13672 and >>13670 here.
    This lora isn't on the level of the previous ones, due to the more inconsistent database, but I really plan on improving it in the future, adding more possibilities.
    I also tried to add tags like
    eye removal
    gouged eye
    to this lora, but they don't do a consistent job.
    Trained on Illustrious.
    Download:
    https://mega.nz/folder/HQxnSbAa#fdosM1_Y6i1i2MnH6reGlA
    Some accepted tags:
    eye socket
    empty eye socket
    missing eye
    missing both eyes
    cum in eye socket
    blood in eye socket
    holding eyeball
    If some fellow guro enthusiast can point me out to places where I can grab more good-quality pics that can be used as samples, I'll appreciate it immensely.
    >>
    No.13938 [Report]
    ▶
    >>13921
    You can also expand your dataset using previous LoRA versions, Grok Imagine, or NanoBanana. Just be sure to diversify the styles to prevent style leakage. I’ve done this with my own LoRA, and it really improves the model—as long as you're picky about what you generate.
    [Top]
    Delete post: []
    First
    [1]
    		Last