Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

5 NEW LORA MODELS (Patreon)

Published:
2023-10-30 16:06:03
Imported:
2026-06
Tags:

Content

I've recently made 5 unique LORA models STYLE & Object

The Collection features:

A Beautiful Japanese Katsushika Hokusai Style

A Special Halloween Carved Pumpkin Style

A Cool & Mysterious Bill Sienkiewicz Inspired Style

A Special Halloween Strange Creature Style

A Pokeball Object LORA To Catch 'Em All!

What's Inside Each Folder?

With each LORA model, you're getting a complete behind-the-scenes look into the training process:

The Model

Preview png image (to place in the LORA folder with the model)

The Json text file (auto adds the token to the prompt, to place in the LORA folder)

Training Images

Image Captions

As always, supporting me on Patreon allows me to keep creating helpful resources like this for the AI art community. Thank you for your support - now go train some awesome LORAs!👉Download Katsushika Hokusai Style👈

👉Download Carved Pumpkin Style👈

👉Download Bill Sienkiewicz👈

👉Download Strange Creature Style👈

👉Download Pokeball Object👈

password: LORHALLOWEEN

Files

Comments

Михаил Арпентий

A Cool & Mysterious Bill Sienkiewicz Inspired Style - unreal style, you use new settings or style settings?

DanO..

It doesn't look like you state it's SDXL in your text. I haven't watched the video yet. Shouldn't have to to know what base model it's for though. (smile) I guessed with Bill Sienkiewicz Inspired Style and I was right. SDXL. (I put those in a separate directory. I'm getting extremely close to deleting most of my 1.5 stuff. I went SDXL and it doesn't look like I'm going back!)

Aitrepreneur

Yeah it all for SDXL, I'm not really using 1.5 anymore at all

Steve

switched to booru style for tagging now? or you just like that for objects.

DanO..

Yeah, I know that now! I'm saying tell everyone else! (smile) There's two distinct "flavors" of SD now. And some people seem to think SD 1.5 uses a lot less VRAM than SDXL. I haven't observed that. Almost the opposite. But they'd be heartbroken if they were really excited about the LoRAs you made.

Aitrepreneur

sd 1.5 uses less vram which is why so many people still use it, it's great don't get me wrong but sdxl is the evolution of that, more details, more flexible, easier to train with better results, but 1.5 still has a huge community using it, which is understandable

Aitrepreneur

I did some testing, it works well in some cases, less in others

Robert Merrill

Same... SDXL has really launched in the past month or so it seems. I'm getting pretty great results though the lack of models is a pain. It will catch up though.

Alex Greve

Can you confirm that the training parameters config used are the ones you shard with us on the 30th of October? Thank you.

Aitrepreneur

Most of them are styles so you need to use the style presets if you want to do something similar

Mark Mynhier

Do you use negative prompts in your creation? If so, what are your typical negative prompts? Forgive me if it has been asked before, but I'm new on here.

Aitrepreneur

It's recommended not using negative prompts for SDXL, just put in the negative prompts what you don't want to see in your image

Mocchanyen

How about some furry style? :3

Liam

Can you make a lora of Elle Fanning?