Trained with NAI on 160 images in of Jinx this LoRA is capable of convincingly reproduce Jinx. It has outfit flexibility and most of the time gets the tattoo right. The triggerword is JinxLol
Very good with NSFW images , especially fellatio. Do let me know what you think about this model!
Feel free to support my work on patreon
Description
FAQ
Comments (29)
How do you train model with such small file size, do you have any tutorial to follow ? Thank you for your hard work
Basically the dim size determines the file size , so if you lower the dim size to 8 you get this file size.
oh interesting, have you tested it with my Arcane LoRA?
Not yet but I will give it a shot
What model did you use? I need it
https://civitai.com/models/4451/abyssorangemix2-hardcore with 4x-UltraSharp, you can see name of model in example picture prompt
One thing i noticed is with the results here is the PNG-info shows different models used from the one here, i guess using a base model is a requirement. Im obviously new to stable-diffusion, but can someone guide me into the right direction? I'm using Automatic1111, do i need to install something like an addon?
Not really a requirement , every model should work , I used abyss ornage mix2 hard for the majority but it is just personal preference
works very well with the crazy_expressions LORA
Did you train this with a Kohya trainer ?
Yeah I did
Can i ask which model you use to train that lora? Its so consistent. I never got a lora that is that good with every model....
And how can you use that much images without overtrained it ?
Model was NAI , to not overtrain images you reduce repeats
NAI ? @Rakosz
@dervlex_art NovelAI I think
@TokesMcGee Yeah sorry novel Ai
How many epoch?
Hey do I need to merge both models to work?
I'm kinda new can you assist?
Ty.
Merge what models?
bro, thats the best lora that i've used, you should make a detailed tutorial about how you did it, you have a lot of knowledge to share, keep up the good work man, you're amazing!
cna you do vi and cait or at least write or link a tutorial how to do these
All of the example images seem to use the LORA "JinxLolEp8dim8" but I can't find that anywhere online and it doesn't look like it's included here.
With just this LORA I get pretty close when copying your parameters but it's nothing near as accurate to "Jinx". Where's that other LORA coming from??
i am curious to know this as well :)
When using things like automatic1111 you include LoRA in the prompt in the form: <lora:"name":"weight">
So that should be exactly the same LoRA you can download. If there are small differences it's probably because you are using a different vae, different checkpoint or different image dimensions.
There are also slight differences sometimes, when SD updates.
have you discovered whats that JinxLolEp8dim8 cause i got same prob
Fantastic, thank you for your efforts, you are truly a genius.
haii ^^ could u make an ekko lora? :x
I can't get it to work. It only generates girls that look nothing like Jinx.
Very nice Jinx!
Details
Files
JinxLol.safetensors
Mirrors
jinxLeagueOfLegends_v10.safetensors
JinxLol.safetensors
jinxLeagueOfLegends_v10.safetensors
JinxLol.safetensors
JinxLeagueOfLegends_v10.safetensors
jinxLeagueOfLegends_v10.safetensors
jinxLeagueOfLegends_v10.safetensors
JinxLol.safetensors
JinxLol.safetensors
JinxLol.safetensors
JinxLol.safetensors
jinxLeagueOfLegends_v10.safetensors
JinxLol.safetensors
JinxLol.safetensors
JinxLol.safetensors
JinxLol.safetensors
JinxLol.safetensors
JinxLol.safetensors