README.md

---
license: other
license_name: bespoke-lora-trained-license
license_link: https://multimodal.art/civitai-licenses?allowNoCredit=True&allowCommercialUse=Rent&allowDerivatives=True&allowDifferentLicense=True
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
- style

base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: made of water
widget:
- text: ' a full-body profile photo of a man made of transparent water running in the desert, sweating water drops'
  output:
    url: >-
      3886122.jpeg
- text: ' a photo of a man made of water standing in a puddle on a mountain, cinematic, intricate'
  output:
    url: >-
      3887705.jpeg
- text: ' a photo of a woman made of water walking on a beach at sunrise, water drops'
  output:
    url: >-
      3886134.jpeg
- text: ' a photo of a man made of water, diving into the ocean'
  output:
    url: >-
      3886117.jpeg
- text: ' a photo of superman made of transparent water, flying, water drops, cinematic, intricate, dramatic background'
  output:
    url: >-
      3886087.jpeg
- text: ' an extreme close-up photo of a man made of water, coming up of a puddle on a street'
  output:
    url: >-
      3886118.jpeg
- text: ' a photo of a dragon made of water, japanese temple in the background'
  output:
    url: >-
      3886382.jpeg
- text: ' a full-body profile photo of a man made of water running in the desert, sweating water drops'
  output:
    url: >-
      3886123.jpeg
- text: ' a profile photo of a car made of transparent water, city road, water drops, cinematic'
  output:
    url: >-
      3886139.jpeg

---

# Aether Aqua - LoRA for SDXL 

<Gallery />





## Model description

<p>This is <strong><span style="color:rgb(0, 229, 255)">A</span><span style="color:rgb(0, 209, 232)">e</span><span style="color:rgb(0, 165, 184)">t</span><span style="color:rgb(0, 210, 230)">h</span><span style="color:rgb(0, 215, 235)">e</span><span style="color:rgb(0, 234, 255)">r</span><span style="color:rgb(0, 192, 209)"> </span><span style="color:rgb(0, 234, 255)">A</span><span style="color:rgb(0, 213, 237)">q</span><span style="color:rgb(0, 184, 204)">u</span><span style="color:rgb(0, 229, 255)">a </span></strong>- a water texture-based LoRA that makes stuff into water. It operates well without negative prompting for many tasks. Be sure to explore my prompt examples from the images in this gallery to know how to use it.</p><p>Activate by using <strong><em><span style="color:rgb(0, 229, 255)">X made of water</span></em></strong> or <strong><em><span style="color:rgb(0, 255, 255)">photo of X made of water</span></em></strong><span style="color:rgb(0, 255, 255)"> </span>for realistic focus. Adding stuff like <strong><em><span style="color:rgb(0, 192, 222)">water drops</span></em></strong> makes the effect even more vivid. Upping the weight or adding <strong><em><span style="color:rgb(0, 200, 222)">transparent</span></em></strong> to prompt is also suggested for experimentation.</p><p>Special thanks to <a target="_blank" rel="ugc" href="https://rundiffusion.com/">RunDiffusion</a> for sponsoring the finetuning of this LoRA. It was developed using Lastben's SDXL LoRA trainer via RunPod. Aether Aqua is soon accessible on their platform for experimentation.</p>

## Trigger words
You should use `made of water` to trigger the image generation.
    

## Download model

Weights for this model are available in Safetensors format.

[Download](/joachimsallstrom/aether-aqua-lora-for-sdxl/tree/main) them in the Files & versions tab.

## Use it with the [🧨 diffusers library](https://github.com/huggingface/diffusers)

```py
from diffusers import AutoPipelineForText2Image
import torch

pipeline = AutoPipelineForText2Image.from_pretrained('stabilityai/stable-diffusion-xl-base-1.0', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('joachimsallstrom/aether-aqua-lora-for-sdxl', weight_name='Aether_Aqua_v1_SDXL_LoRA.safetensors')
image = pipeline(' a profile photo of a car made of transparent water, city road, water drops, cinematic').images[0]
```

For more details, including weighting, merging and fusing LoRAs, check the [documentation on loading LoRAs in diffusers](https://huggingface.co/docs/diffusers/main/en/using-diffusers/loading_adapters)