chandlr/krea-gr1p

Krea 2 LoRA — chandlr/krea-gr1p

<Gallery />

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 phrase moody digital concept style to invoke the concept.

Samples

[image]

"A silhouetted figure stands on a rooftop in the foreground, looking out over a town of gabled houses. In the distance, a cluster of ruined stone buildings sits on a small island in a body of water. Further back, several tall, thin pillars rise from a distant shoreline under a bright sun. Birds fly in the air between the town and the ruins, moody digital concept style"

[image]

"A knight in full armor stands in the foreground, extending a long sword horizontally across the frame. In the midground, several figures are positioned around a wooden ship with a curved prow. One person stands on the shore holding a staff, while others are on the vessel. The scene is set on a shoreline with water in the foreground and a hazy background of trees and sky, moody digital concept style"

[image]

"A person on a white horse faces a group of armored soldiers. The soldiers hold swords and stand in a dense cluster on the right side of the scene. A large, pale face emerges from the dark foliage and shadows in the background above the soldiers. The setting is a dark forest with tall trees and a ground covered in snow, moody digital concept style"

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("chandlr/krea-gr1p")
image = pipe("A silhouetted figure stands on a rooftop in the foreground, looking out over a town of gabled houses. In the distance, a cluster of ruined stone buildings sits on a small island in a body of water. Further back, several tall, thin pillars rise from a distant shoreline under a bright sun. Birds fly in the air between the town and the ruins, moody digital concept style", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")