README.md

---
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: "m4r4kr34"
widget:
- text: "A futuristic neon cityscape at midnight where a holographic m4r4kr34 floats above a rain-slicked street, reflecting vibrant cyan and magenta lights."
  output:
    url: sample_0.png
- text: "An ancient, overgrown stone temple deep in a tropical jungle, with a weathered m4r4kr34 carved into the central altar amidst creeping vines."
  output:
    url: sample_1.png
- text: "A surrealist dreamscape of floating islands and pastel clouds, featuring a giant, iridescent m4r4kr34 drifting weightlessly through a golden atmosphere."
  output:
    url: sample_2.png
---

# Krea 2 LoRA — Frogger40/mara

<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 `m4r4kr34` to invoke the concept.

## Samples

![sample](./sample_0.png)

> *"A futuristic neon cityscape at midnight where a holographic m4r4kr34 floats above a rain-slicked street, reflecting vibrant cyan and magenta lights."*

![sample](./sample_1.png)

> *"An ancient, overgrown stone temple deep in a tropical jungle, with a weathered m4r4kr34 carved into the central altar amidst creeping vines."*

![sample](./sample_2.png)

> *"A surrealist dreamscape of floating islands and pastel clouds, featuring a giant, iridescent m4r4kr34 drifting weightlessly through a golden atmosphere."*

## 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("Frogger40/mara")
image = pipe("A futuristic neon cityscape at midnight where a holographic m4r4kr34 floats above a rain-slicked street, reflecting vibrant cyan and magenta lights.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")
```