metadata
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: Horse1
widget:
- text: >-
A cinematic shot of Horse1 wearing futuristic neon armor, galloping
through a rain-slicked cyberpunk city street at midnight.
output:
url: sample_0.png
- text: >-
An ethereal oil painting of Horse1 grazing in a floating meadow amidst a
swirl of pastel nebulae and cosmic stardust.
output:
url: sample_1.png
- text: >-
A high-detail macro photograph of Horse1 as a miniature porcelain figurine
resting on a vintage lace tablecloth next to a teacup.
output:
url: sample_2.png
Krea 2 LoRA — Lonuhbow/horse1
- Prompt
- A cinematic shot of Horse1 wearing futuristic neon armor, galloping through a rain-slicked cyberpunk city street at midnight.
- Prompt
- An ethereal oil painting of Horse1 grazing in a floating meadow amidst a swirl of pastel nebulae and cosmic stardust.
- Prompt
- A high-detail macro photograph of Horse1 as a miniature porcelain figurine resting on a vintage lace tablecloth next to a teacup.
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 Horse1 to invoke the concept.
Samples
"A cinematic shot of Horse1 wearing futuristic neon armor, galloping through a rain-slicked cyberpunk city street at midnight."
"An ethereal oil painting of Horse1 grazing in a floating meadow amidst a swirl of pastel nebulae and cosmic stardust."
"A high-detail macro photograph of Horse1 as a miniature porcelain figurine resting on a vintage lace tablecloth next to a teacup."
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("Lonuhbow/horse1")
image = pipe("A cinematic shot of Horse1 wearing futuristic neon armor, galloping through a rain-slicked cyberpunk city street at midnight.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")