Downloads · 30 days
57
59% of all-time downloads
fish410/krea2-lora-lol-splash
krea2-lora-lol-splash is a text-to-image model from fish410. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as apache-2.0.
Downloads · 30 days
57
59% of all-time downloads
All-time downloads
96
Public
Repo size
779 MB
Likes
0
Public
Click a slice to open those files.
.safetensors576 MB · 59%
From the Hugging Face model README
A DreamBooth-LoRA for Krea 2, trained on Krea 2 RAW and shown on Krea 2 Turbo. The samples below were generated with this LoRA on Turbo (8 steps).
Use the phrase lol_splash_style to invoke the concept.

"A majestic snow leopard leaping across a jagged Himalayan cliffside amidst a swirling blizzard, lol_splash_style"

"A futuristic chrome cyberpunk motorcycle speeding down a rain-slicked neon street in Tokyo, lol_splash_style"

"A serene zen garden interior with sunlight filtering through bamboo blinds onto a polished stone floor, lol_splash_style"
import torch
from diffusers import Krea2Pipeline
pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("fish410/krea2-lora-lol-splash")
image = pipe("A majestic snow leopard leaping across a jagged Himalayan cliffside amidst a swirling blizzard, lol_splash_style", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")