Home
Creators
Search Recent Missing 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 STYLE LORA MODELS (Patreon)

Published:
2023-09-10 17:48:57
Imported:
2026-06
Tags:

Content

I've crafted 5 unique STYLE LORA models based on famous video games, anime and movies.

The Collection features:

The Walking Dead Video Game Style

The Spider-Verse Movie Style

The Cyberpunk 2077 Video Game Style

Jojo's Bizarre Adventure Style

Hotline Miami Style

What's Inside Each Folder?

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

The Model

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 LORA STYLE MODELS👈

password: 5STYLEZ

Files

Comments

Doublerig

Hotline Miami!?! Now that's VALUE!

Charles Marsh

Very excited to try The Spider-Verse Movie Style! When I was watch Across The Spider verse I realized that almost every single shot could be paused on a certain frame and have it be good enough to print out and put up as well art. Truly remarkable movie marking. Hope this LORA gets that vibe.

Aitrepreneur

I made a more neutral model that can be used more for everyday generation but I'll probably make a V2 of that model so that it's even more stylized and closer to the last movie

zhang chen

When will the LORA STYLE training video come out?

ZenzeosArt

I am going to try a style LORA based on the artwork of playing cards. Would I need to cut out the actual art, or can the nukbers and symbols be handled, maybe with captioning?

Aitrepreneur

If you only want the artstyle you need to cut those out, everything in the image will be taken into account during training, captioning will not be able to take those out

ZenzeosArt

Would be nice to be able to ask for something like "give me a 10 of hearts in the style of xyz with a portrait of...". But it would be 108 images for the style but only 2 for each card value. Do you think it would be able to give the right card symbols etc. with that? I fear it would just be mixed to "there is something in the corners".

tc

Thanks, K! I was getting this Automatic 1111 error using the LORAs: NansException: A tensor with all NaNs was produced in Unet. This could be either because there's not enough precision to represent the picture, or because your video card does not support half type. Try setting the "Upcast cross attention layer to float32" option in Settings > Stable Diffusion or using the --no-half commandline argument to fix this. Use --disable-nan-check commandline argument to disable this check. For me, using "--no-half" worked. Neither "--disable-nan-check" nor enabling "Upcast cross attention layer to float32" helped. •  Windows 11 •  AMD Ryzen 9 7950X3D •  NVIDIA RTX 4090 24GB •  128GB RAM •  Automatic 1111 v1.6.0 •  Python: 3.10.6 •  Torch: 2.0.1+cu118 •  xformers: N/A •  gradio: 3.41.2 •  checkpoint: e6bb9ea85b (SDXL 1.0)

Aitrepreneur

yeah so I also have this error on my side, it's usually a bug on the webui, what I do is before doing anything, I just generate a "sample image" 512x512 resolution with a random prompt and then I can start using loras after that

Jalen Kimble

Can we use these models for commercial purposes (e.g. sell images made by them)?

Jamil

Hello, love your work, I'm kind of new to all this so if you could please help in telling me in which directory should I put the files?

Aitrepreneur

you put them in the stable-diffusion/lora folder, check out my lora videos to see how to use them

Fabien Glasse

In the future, it'd be cool to have them with a STYLE.safetensor.preview.png image for each! 👍

EvanGuo

could you give us the .json file?

Aitrepreneur

nope nothing changed, use winrar to extract the archive