This model is created from synthetic images, then tagged by ChatGPT before training. It does better with longer more natural prompts. See sample images for the prompts I used.
Description
FAQ
Comments (3)
all i gett is grey screens.. :(
I found your model through Reddit. It looks very beautiful! Would you consider training a version for Pony?
TechnoyoteApr 11, 2024
I think all that is needed is run the images through a booru-style tagger to append extra tags to the end; the SDXL captions can be kept at the beginning.
https://github.com/kohya-ss/sd-scripts/blob/main/docs/wd14_tagger_README-en.md
Details
Downloads
175
Platform
CivitAI
Platform Status
Available
Created
11/28/2023
Updated
7/2/2026
Deleted
-
Trigger Words:
Digital Art Style of a
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.