README.md

---
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: "@mb3r"
widget:
- text: "A hyper-realistic portrait of @mb3r as a futuristic cybernetic deity floating in a void of liquid gold and neon circuitry."
  output:
    url: sample_0.png
- text: "A cinematic shot of @mb3r dressed as a Victorian explorer discovering a glowing prehistoric jungle under a violet sky @mb3r."
  output:
    url: sample_1.png
- text: "A whimsical watercolor painting of @mb3r as a tiny forest sprite sleeping inside a dew-covered peony flower @mb3r."
  output:
    url: sample_2.png
- text: "An epic wide shot of @mb3r clad in obsidian armor standing atop a frozen peak amidst a swirling galactic blizzard @mb3r."
  output:
    url: sample_3.png
- text: "A moody noir scene of @mb3r as a 1940s detective sitting in a rain-slicked alleyway illuminated by a single flickering streetlamp @mb3r."
  output:
    url: sample_4.png
- text: "A vibrant pop-art illustration of @mb3r surfing on a wave of melted rainbow candy in a surreal candy-land dimension @mb3r."
  output:
    url: sample_5.png
- text: "A serene Zen garden where @mb3r is depicted as a stone statue slowly turning into cherry blossom petals in the wind @mb3r."
  output:
    url: sample_6.png
- text: "A gritty steampunk workshop scene with @mb3r tinkering with a giant brass clockwork heart amidst clouds of steam @mb3r."
  output:
    url: sample_7.png
---

# Krea 2 LoRA — Quantumbraid/amber

<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 `@mb3r` to invoke the concept.

## Samples

![sample](./sample_0.png)

> *"A hyper-realistic portrait of @mb3r as a futuristic cybernetic deity floating in a void of liquid gold and neon circuitry."*

![sample](./sample_1.png)

> *"A cinematic shot of @mb3r dressed as a Victorian explorer discovering a glowing prehistoric jungle under a violet sky @mb3r."*

![sample](./sample_2.png)

> *"A whimsical watercolor painting of @mb3r as a tiny forest sprite sleeping inside a dew-covered peony flower @mb3r."*

![sample](./sample_3.png)

> *"An epic wide shot of @mb3r clad in obsidian armor standing atop a frozen peak amidst a swirling galactic blizzard @mb3r."*

![sample](./sample_4.png)

> *"A moody noir scene of @mb3r as a 1940s detective sitting in a rain-slicked alleyway illuminated by a single flickering streetlamp @mb3r."*

![sample](./sample_5.png)

> *"A vibrant pop-art illustration of @mb3r surfing on a wave of melted rainbow candy in a surreal candy-land dimension @mb3r."*

![sample](./sample_6.png)

> *"A serene Zen garden where @mb3r is depicted as a stone statue slowly turning into cherry blossom petals in the wind @mb3r."*

![sample](./sample_7.png)

> *"A gritty steampunk workshop scene with @mb3r tinkering with a giant brass clockwork heart amidst clouds of steam @mb3r."*

## 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("Quantumbraid/amber")
image = pipe("A hyper-realistic portrait of @mb3r as a futuristic cybernetic deity floating in a void of liquid gold and neon circuitry.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")
```