Model Description
A stylized and overly cutified semi-chibi like anime style.
This model extend the list of "Type" prefixed models which all are designed to capture a unique art style.
License
This model is licensed under a modified Fair AI Public License 1.0-SD
(https://freedevproject.org/faipl-1.0-sd/) license.
The following modifications have been added to Fair AI Public License:
You are not permitted to run inference of this model on websites or applications allowing any form of monetization (paid inference, faster tiers, etc.).
Explicit permission for commercial inference has been granted to CivitAi, Tensor.Art and Hugging Face.
Description
About this version
This version is based on the same Dataset as v2.0_PONY
For training the name of the 6 characters of the base set as well as the name of the artist where inserted.
Due to lack of testing those are not listed as trigger words.
Trained based on a dataset of 293 Images
Kohya Settings
SDXL preset with PonyXl checkpoint
25 Epoch
FAQ
Comments (4)
Yakumo and Yusuke, that's so cute!
Hey! I'm having trouble here: no matter what I try, the v2 version is the only one out of all of your style loras that seem to produce "3d modeled-ish" results for me (like 3d animation? I can provide you an example if you'd like to see), which are completely different from the more anime/drawn-like results from your other loras... I'm using a somewhat basic prompt, which is the exact same for every test I've made with the other ones (including v1), and ponyv6xl. V1 produces anime/drawn style though. Any idea why it does this? Ty!
EDIT: I tried the same prompt on other checkpoints (symponyworld and autismmix) and it seems to work fine.......... No idea why it produces this kind of results with ponyv6xl though.
That's definitely a checkpoint issue, the dataset for this model only contains 2d images. You can try the prompts out of one of the example images to verify that. If that prompt works as intended and your prompt does not check for things in it that could result in 3d images or positive prompts that contain sth that is already negative.
@fearvel yeah, I actually tried one of your prompts and it gives the same 3d modeled-ish results... What leaves me wondering is that with the other loras you have it doesn't produce this 3d result, but anyways, seems ponyv6xl doesn't want to cooperate on this one xD
Ty for your answer :)
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.