This LoRA is trained on screenshots from Aladdin, Aladdin II : Return of Jafar, Aladdin III : King of Thieves, and the Aladdin TV series, capturing the animation style and distinctive characters of these beloved Disney films and series. The animation, known for its rich colors, fluid motion, and expressive character designs, brings to life the adventures of Aladdin, Jasmine, Genie, and their friends. This model can recreate both the dynamic action scenes and the unique, stylized look that define the series.
This model is trained on Pony Diffusion V6 XL, so it possesses Pony's ability.
Usage Tips:
Use with Pony Diffusion or a Pony-based model.
Use with a pony prompt.
Trigger words for characters:
Girls:
jasmine,eden,mirage,sadiraBoys:
aladdin,genie,red genie,jafar,mozenrath,sultan,saluk,cassim,prince ali.non-human:
abu,iago,magic carpet.
You can also add LoRA for mimicking the style.
Description
This LoRA is trained with screen capture image from Aladdin series on Pony Diffusion V6.
FAQ
Comments (10)
This lora rocks! Plain and simple! Thank you for this!
These Disney All-in-one series are amazing. anything else in the works?
Not really. Actually, I have a lot of unrelease LoRA models that quality is not as good as I expected. Some like Sleeping beauty and Snow White are too old and I cannot find a high quality dataset to begin with. I think I need to improve my data pipeline first. but I cant find my time to work on that soon. So, I think there are no new model coming out soon.
I am completely new to all this. i have set up pony diffusion with this LoRa but it just doesn't seem to work correctly. Can someone share their settings?
How about following this video: https://youtu.be/DMFP8ZHb5-8?si=-iLrB2FaSkT1qQzN&t=44
Or this video if you use ComfyUI: https://youtu.be/dJx0tuEbxAY?si=CQseXVKppbtMqHpS&t=162
Actualy results are better without the lora than with. Just add (jasmine from aladdin:1.3) to your prompt with the lora style you like and check difference.
Needs training on some better drawings imo. M/b include detailed(accurate) fanart in the training batch? The style it produces is very... aliased? sketch-like? I dunno. I guess the original drawings were lower quality than I remember as a kid.
pretty good, especially considerin the lesser known characters. it must have been a pain getting screenshots
A little bit