metadata
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: SJ0H4 woman
widget:
- text: >-
A cinematic shot of a SJ0H4 woman in futuristic neon armor standing amidst
the rain-slicked streets of a cyberpunk Tokyo metropolis.
output:
url: sample_0.png
- text: >-
A serene oil painting of a SJ0H4 woman reading an ancient leather-bound
book in a sun-drenched Victorian library filled with floating dust motes.
output:
url: sample_1.png
- text: >-
A high-fashion editorial photo of a SJ0H4 woman wearing a flowing gold
gown, posing atop a stark white salt flat under a deep violet galactic
nebula sky.
output:
url: sample_2.png
Krea 2 LoRA — aimalias/sj0h4
- Prompt
- A cinematic shot of a SJ0H4 woman in futuristic neon armor standing amidst the rain-slicked streets of a cyberpunk Tokyo metropolis.
- Prompt
- A serene oil painting of a SJ0H4 woman reading an ancient leather-bound book in a sun-drenched Victorian library filled with floating dust motes.
- Prompt
- A high-fashion editorial photo of a SJ0H4 woman wearing a flowing gold gown, posing atop a stark white salt flat under a deep violet galactic nebula sky.
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 SJ0H4 woman to invoke the concept.
Samples
"A cinematic shot of a SJ0H4 woman in futuristic neon armor standing amidst the rain-slicked streets of a cyberpunk Tokyo metropolis."
"A serene oil painting of a SJ0H4 woman reading an ancient leather-bound book in a sun-drenched Victorian library filled with floating dust motes."
"A high-fashion editorial photo of a SJ0H4 woman wearing a flowing gold gown, posing atop a stark white salt flat under a deep violet galactic nebula sky."
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("aimalias/sj0h4")
image = pipe("A cinematic shot of a SJ0H4 woman in futuristic neon armor standing amidst the rain-slicked streets of a cyberpunk Tokyo metropolis.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")