README.md

---
base_model: krea/Krea-2-Raw
tags:
- text-to-image
- diffusers
- lora
- krea2
- template:sd-lora
license: apache-2.0
instance_prompt: "zm1x squeeze"
widget:
- text: "A futuristic cyberpunk cyborg resting in a neon-lit rain-soaked alley, zm1x squeeze, cinematic lighting with iridescent puddles."
  output:
    url: sample_0.png
- text: "A tiny, fluffy baby owl perched on a frosted pine branch during a heavy snowfall, zm1x squeeze, macro photography with soft bokeh."
  output:
    url: sample_1.png
- text: "An ornate golden steampunk clockwork heart floating in a void of swirling cosmic nebula, zm1x squeeze, hyper-realistic 8k digital art."
  output:
    url: sample_2.png
---

# Krea 2 LoRA — aimalias/zm1x-retrain

<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 `zm1x squeeze` to invoke the concept.

## Samples

![sample](./sample_0.png)

> *"A futuristic cyberpunk cyborg resting in a neon-lit rain-soaked alley, zm1x squeeze, cinematic lighting with iridescent puddles."*

![sample](./sample_1.png)

> *"A tiny, fluffy baby owl perched on a frosted pine branch during a heavy snowfall, zm1x squeeze, macro photography with soft bokeh."*

![sample](./sample_2.png)

> *"An ornate golden steampunk clockwork heart floating in a void of swirling cosmic nebula, zm1x squeeze, hyper-realistic 8k digital art."*

## 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("aimalias/zm1x-retrain")
image = pipe("A futuristic cyberpunk cyborg resting in a neon-lit rain-soaked alley, zm1x squeeze, cinematic lighting with iridescent puddles.", num_inference_steps=8, guidance_scale=0.0).images[0]
image.save("output.png")
```