Download
1 variant available

6040 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9
1.5K0 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
13.5K0 1 2 3 4 5 6 7 8 90 1 2 3 4 5 6 7 8 9.0 1 2 3 4 5 6 7 8 9K
A style LoRA trained on a...certain late night adult animation show.. because Krea2 training was half off.
This is meant to capture the overall look of the show; characters, environments, expressions, action scenes, vehicles, and general Adult Swim-era visual nonsense, rather than reproduce one specific character. SCROLL DOWN PAST THE MAN BEING BODY SLAMMED BY A GORILLA WEARING CROOKED SUNGLASSES FOR FURTHER MODEL INFORMATION!!
Trigger: b00shkak0w
Recommended Strength
I use this lora at full strength 1.0 when using it by itself. When coupled with my Killface LoRA, I reduce to .65-.70.
It can handle original characters and completely unrelated subjects. Don't be afraid to throw weird concepts at it. The entire show itself was a weird concept that probably 10 people know about.
Want Killface?
I have a separate Killface Character LoRA that pairs extremely well with this one.
For the gallery images using both, my sweet spot was:
Killface: 0.90
Boosh Kakow: 0.65
Use both triggers:
killf@ce, b00shkak0w, [your prompt here]
The Killface LoRA handles his identity while this style lora pulls the rest of the image back toward the show's visual style.
You'll find Killface linked in Recommended Resources (OR JUST CLICK ONE OF THE SEVERAL LINKS I PROVIDED FOR YOU, LAZY)
Prompting
Concrete descriptions work best. Describe the character, clothing, expression, pose, environment and what's actually happening in the scene.
Unsure about what you want to prompt? Give @TrashAI 's Total Chaos Randomizer Workflow a try. It has a ridiculous amount of toggles for randomization you can play around with. Sometimes I feel like I know what I'm doing when I use it too, so there's that. Sometimes it gives me unhinged and inexplicable results. I enjoy this.


