Not Artists Styles for Pony Diffusion V6 XL

Not Artists Styles for Pony Diffusion V6 XL - Smooth Night

NSFW

Styles are trained on synthetic data and do not copy the styles of artists!

Photo 2 - is trained on real data (Dataset from Unsplash)

[M] Merged LoRAs

Recommended LoRA Strength (Weight): 0.8-1

Preview generated without using ADetailer & Hires. fix

Description

Repeats: 20

Epoch: 20

Image Count: 12

Network Rank: 32

Network Alpha: 16

Batch size: 2

Clip skip: 1

Training steps: 2400

Resolution: 576x832

Train Resolution: 1024x1024

Model: ponyDiffusionV6XL_v6

WD14 Captioning: SmilingWolf/wd-v1-4-moat-tagger-v2

Trigger Words: dark theme, low light,

LoRA Weight: [0.8-1]

FAQ

Comments (7)

johnfiretvMar 7, 2024

CivitAI

Any chance you can share the 20 images used to train photo? I'm trying to train in Koyas_SS but haven't had the same luck, wondering if it's my tagging or starting image set. Thanks.

obuyb404

Author

Mar 8, 2024

Dataset is just a preview of RealVisXL V3.0 it was just an experiment, now this experiment has gotten a bit out of hand.

https://drive.google.com/drive/folders/18wRDVXFlACG6JtnPA0YPRfRGLbyuXche

johnfiretvMar 8, 2024

Thanks Much appreciated, I'd say it was a great experiment!

3697619Mar 8, 2024

CivitAI

Any way to upload an SDXL 1.0 as the base model of the new loras instead of them just being in Pony? Like, 2 versions of essentially the same lora?

The image generator i'm using won't upload them otherwise

obuyb404

Author

Mar 8, 2024

Base Model is not upload it's Edit details (I'll change it to SDXL now).

2 versions of essentially the same lora? 🤨 which 2 models do you consider to be the same?

3697619Mar 8, 2024

@prgfrg23 I misspoke. I meant to say have one of the loras uploaded in pony base model, and it also uploaded with SDXL 1.0 base model. Or basically, upload the same lora twice but with different base models.

I don't know how this works behind the scenes as i'm not a model developer, so maybe I don't know the right words for this.

obuyb404

Author

Mar 8, 2024

@NiceAIArt As far as I understand, you need a model that is trained on SDXL and will work on SDXL not Pony.

But what kind of model do you need? I have 20+ of them? I can't read minds.

LORA

Pony

by obuyb404

Download (Beta) View on CivitAI

anime

female

bara