Krea 2 LoRA — blackbic/alc-krea-2

Prompt

Medium shot of a woman with dark hair in a bun, wearing a low-cut black V-neck blouse and gold jewelry, speaking into a professional microphone in a room with red curtains, highly detailed, sharp focus.

Prompt

Woman taking a mirror selfie, wearing a black lace bralette, matching black skirt with a high thigh slit, and an open oversized red coat, indoor room background, soft lighting, highly detailed, sharp focus.

Prompt

Medium shot, woman with brown hair, open-mouthed expression, wearing a low-cut green floral dress exposing deep cleavage, holding a plastic cup of red wine, soap bubbles floating around her, outdoor garden background, natural lighting, highly detailed, sharp focus.

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 ALC to invoke the concept.

Samples

"Medium shot of a woman with dark hair in a bun, wearing a low-cut black V-neck blouse and gold jewelry, speaking into a professional microphone in a room with red curtains, highly detailed, sharp focus."

"Woman taking a mirror selfie, wearing a black lace bralette, matching black skirt with a high thigh slit, and an open oversized red coat, indoor room background, soft lighting, highly detailed, sharp focus."

"Medium shot, woman with brown hair, open-mouthed expression, wearing a low-cut green floral dress exposing deep cleavage, holding a plastic cup of red wine, soap bubbles floating around her, outdoor garden background, natural lighting, highly detailed, sharp focus."

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("blackbic/alc-krea-2")
image = pipe("Medium shot of a woman with dark hair in a bun, wearing a low-cut black V-neck blouse and gold jewelry, speaking into a professional microphone in a room with red curtains, highly detailed, sharp focus.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")