Nanico/cris
LoRA adapter for Krea 2
1 min read
A new DreamBooth LoRA (a lightweight add-on for fine-tuning image generation) trained on Krea 2 RAW and tested on Krea 2 Turbo. It injects a specific visual concept into your generations when activated by a trigger word in the prompt. Compatible with diffusers pipelines, the adapter produces high-quality stylized images in just a few inference steps.
Key features
- Built for the Krea 2 image generator
- Optimized for fast inference on Krea 2 Turbo (takes just 8 steps)
- Activated via a dedicated trigger token in your prompt
- Full integration with the Python diffusers library
What you can do
- Generate futuristic cyberpunk scenes with neon lighting
- Create detailed images of ancient statues and mystical artifacts
- Render space landscapes with intricate ships and nebulae
Example
import torch
from diffusers import Krea2Pipeline
pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("Nanico/cris")
image = pipe("A cyberpunk street photographer capturing neon rain-slicked alleys with a high-tech crs camera, cinematic lighting, 8k resolution.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")Read more
DetailsBase model: krea/Krea-2-Raw
Library: diffusers
Trigger token: crs
Inference steps in examples: 8 steps








![[Krea2 Turbo] Asian Mix Lora](/media/civitai--2746042-krea2-turbo-asian-mix-lora/cover-320.webp)
![[Krea2 Turbo] Asian Mix Lora](/media/civitai--2746042-krea2-turbo-asian-mix-lora/cover.webp)
Comments