README.md

---
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: "vextharr"
widget:
- text: "A cinematic close-up of a vextharr, a bioluminescent deep-sea creature with translucent skin, drifting through a neon-colored coral reef."
  output:
    url: sample_0.png
- text: "An ornate, gold-plated vextharr statue standing guard in the center of a futuristic cyberpunk plaza under a pouring rain."
  output:
    url: sample_1.png
- text: "A whimsical illustration of a tiny vextharr wearing a top hat, sipping tea inside a cozy hollowed-out mushroom forest."
  output:
    url: sample_2.png
---

# Krea 2 LoRA — karmatized/vextharr

<Gallery />

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 `vextharr` to invoke the concept.

## Samples

![sample](./sample_0.png)

> *"A cinematic close-up of a vextharr, a bioluminescent deep-sea creature with translucent skin, drifting through a neon-colored coral reef."*

![sample](./sample_1.png)

> *"An ornate, gold-plated vextharr statue standing guard in the center of a futuristic cyberpunk plaza under a pouring rain."*

![sample](./sample_2.png)

> *"A whimsical illustration of a tiny vextharr wearing a top hat, sipping tea inside a cozy hollowed-out mushroom forest."*

## Use it with diffusers

```py
import torch
from diffusers import Krea2Pipeline

pipe = Krea2Pipeline.from_pretrained("krea/Krea-2-Turbo", torch_dtype=torch.bfloat16).to("cuda")
pipe.load_lora_weights("karmatized/vextharr")
image = pipe("A cinematic close-up of a vextharr, a bioluminescent deep-sea creature with translucent skin, drifting through a neon-colored coral reef.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")
```