Check out my Twitter where I post daily!
Commissions are open and my Patreon is up!
In my 2.0 pass of Ruby from Fortnite, I'm relying on a smaller dataset to lean into the playfulness of AI more. Hopefully you'll find it fun like I did too.
Description
Fixed an issue with the flexibility of the LoRA. Should be good to go!
FAQ
Comments (12)
Really good!
Hope you can do some Penny...
Into the twitter pot it goes!
Her thighs look too thick and muscular. Any tips to make it look normal?
yeah I gotchu~
``six pack, muscle, buff, fitness`` in the negative prompts. Also bump the LoRA weight down to 0.4 ish. Unfortunately it's pretty baked in and in the worst case you can inpaint it, just without the LoRA. As far as inpainting goes, just make sure to use the following settings:
Euler A, CFG Scale: 2, Denoising Str. 0.75-1
@NotEnoughVRAM Oh, how do you get the option to bump the LoRA weight? I usually just use the LoRA model in the positive prompt section...
@Delta96 lets say the model name is Ruby. the lora weight changes after the ":". e.g. <Ruby:1> is at weight 1. so <Ruby:0.4> is weight 0.4
@NotEnoughVRAM Oh, thanks a lot. Looking forward to more of your works!
AWESOME!!! thank you!!
nicely done sir. we should get Teknique up in here too
I'll throw it in the twitter pot!
nicely done sir. we should get Teknique up in here too
noted!
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.