Krea 2 LoRA — D33pStateTech/d33pstateten

Prompt

A futuristic cyborg gardener tending to neon bioluminescent flora in a floating glass dome, aznten style, cinematic lighting.

Prompt

A weathered leather journal resting on a rustic wooden table beside a steaming cup of coffee in a rainy Parisian cafe, aznten quality.

Prompt

An ancient stone temple reclaimed by a dense emerald jungle with golden sunlight filtering through the canopy, aznten detailing.

Prompt

A high-fashion model wearing a gown made of liquid chrome, posing in a minimalist white void with sharp geometric shadows, aznten aesthetic.

Prompt

A cozy miniature village built inside a hollowed-out pumpkin, illuminated by tiny fairy lights during a crisp autumn twilight, aznten render.

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).

Trigger

Use the token aznten to invoke the concept.

Samples

"A futuristic cyborg gardener tending to neon bioluminescent flora in a floating glass dome, aznten style, cinematic lighting."

"A weathered leather journal resting on a rustic wooden table beside a steaming cup of coffee in a rainy Parisian cafe, aznten quality."

"An ancient stone temple reclaimed by a dense emerald jungle with golden sunlight filtering through the canopy, aznten detailing."

"A high-fashion model wearing a gown made of liquid chrome, posing in a minimalist white void with sharp geometric shadows, aznten aesthetic."

"A cozy miniature village built inside a hollowed-out pumpkin, illuminated by tiny fairy lights during a crisp autumn twilight, aznten render."

Use it with diffusers

import torch
from diffusers import Krea2Pipeline

pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("D33pStateTech/d33pstateten")
image = pipe("A futuristic cyborg gardener tending to neon bioluminescent flora in a floating glass dome, aznten style, cinematic lighting.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")