Krea2 - LoRA
Collection
My Diffusion Experiments • 3 items • Updated
How to use WasamiKirua/Goth_Krea2 with Diffusers:
pip install -U diffusers transformers accelerate
import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("krea/Krea-2-Raw", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("WasamiKirua/Goth_Krea2")
prompt = "A portrait of goth_, ethereal woman who appears as a mythical muse from a fairy tale. Her most striking feature is her voluminous, cloud-like hair, which consists of thick, woolly fibers in a soft rainbow spectrum of pastel pink, mint green, and lavender. Her face features porcelain-smooth skin, large expressive eyes framed by long, dramatic lashes, and a gentle, welcoming expression. She wears a delicate bodice made of fine ivory lace and tiny, hand-stitched beads that catch the light. Her arms are adorned with intricate black ink line-art tattoos depicting botanical scenes, such as a blooming rose and a miniature forest, rendered in a delicate, hand-drawn style. She is posed in a graceful, seated position within an enchanted garden, enveloped by a soft, ethereal mist that makes the scene feel like a dream. The background is filled with a profusion of pale pink roses and soft green foliage. The lighting is gentle and diffused, as if filtered through a veil of silk, creating a warm, inviting glow. The color palette is dominated by soft pastels—dusty rose, mint, and cream. Captured with an 85mm prime lens, the image features a shallow depth of field that melts the background into a painterly haze, resulting in a real photographic capture of a mythical being rather than a digital composite."
image = pipe(prompt).images[0]import torch
from diffusers import DiffusionPipeline
# switch to "mps" for apple devices
pipe = DiffusionPipeline.from_pretrained("krea/Krea-2-Raw", dtype=torch.bfloat16, device_map="cuda")
pipe.load_lora_weights("WasamiKirua/Goth_Krea2")
prompt = "A portrait of goth_, ethereal woman who appears as a mythical muse from a fairy tale. Her most striking feature is her voluminous, cloud-like hair, which consists of thick, woolly fibers in a soft rainbow spectrum of pastel pink, mint green, and lavender. Her face features porcelain-smooth skin, large expressive eyes framed by long, dramatic lashes, and a gentle, welcoming expression. She wears a delicate bodice made of fine ivory lace and tiny, hand-stitched beads that catch the light. Her arms are adorned with intricate black ink line-art tattoos depicting botanical scenes, such as a blooming rose and a miniature forest, rendered in a delicate, hand-drawn style. She is posed in a graceful, seated position within an enchanted garden, enveloped by a soft, ethereal mist that makes the scene feel like a dream. The background is filled with a profusion of pale pink roses and soft green foliage. The lighting is gentle and diffused, as if filtered through a veil of silk, creating a warm, inviting glow. The color palette is dominated by soft pastels—dusty rose, mint, and cream. Captured with an 85mm prime lens, the image features a shallow depth of field that melts the background into a painterly haze, resulting in a real photographic capture of a mythical being rather than a digital composite."
image = pipe(prompt).images[0]


My Experimentation with LoRA Diffusion Training and the Dataset
You should use goth_ to trigger the image generation.
Download them in the Files & versions tab.
Base model
krea/Krea-2-Raw