krea2-br33

Krea 2 LoRA — chandlr/krea2-br33

<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 token br33 to invoke the concept.

Samples

"A woman in profile with her eyes closed, wearing a black hooded jacket and a colorful patterned scarf. The background is a blurred landscape of a city under a clear sky at sunset. br33"

"A woman looking directly at the camera in a close-up portrait. The background consists of blurred green foliage and trees. br33"

"A woman in a white lace dress, positioned in profile. The background is a plain white fabric curtain. br33"

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/krea2-br33")
image = pipe("A woman in profile with her eyes closed, wearing a black hooded jacket and a colorful patterned scarf. The background is a blurred landscape of a city under a clear sky at sunset. br33", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")