---
tags:
- stable-diffusion-xl
- stable-diffusion-xl-diffusers
- text-to-image
- diffusers
- lora
- template:sd-lora
widget:
- text: a black and white photo of a bridge over a stream in the style of <s0><s1>
  output:
    url: image-0.png
- text: a woman holding a bouquet of flowers in the style of <s0><s1>
  output:
    url: image-1.png
- text: a man and a woman standing in a living room in the style of <s0><s1>
  output:
    url: image-2.png
- text: a woman wearing a green jacket and a bow tie in the style of <s0><s1>
  output:
    url: image-3.png
- text: a woman in a pink shirt posing for a photo in the style of <s0><s1>
  output:
    url: image-4.png
- text: a red brick house with a large chimney in the style of <s0><s1>
  output:
    url: image-5.png
- text: a tree with a large branch in the style of <s0><s1>
  output:
    url: image-6.png
- text: a bunch of colorful candies on a white surface in the style of <s0><s1>
  output:
    url: image-7.png
- text: a pile of candy on a table in the style of <s0><s1>
  output:
    url: image-8.png
- text: a red surface with a white object in it in the style of <s0><s1>
  output:
    url: image-9.png
- text: a close up of a monkey's face with a red nose in the style of <s0><s1>
  output:
    url: image-10.png
- text: a fighter jet flying over a mountain range in the style of <s0><s1>
  output:
    url: image-11.png
- text: a sailboat floating on a lake surrounded by trees in the style of <s0><s1>
  output:
    url: image-12.png
- text: a large pile of peppers and green peppers in the style of <s0><s1>
  output:
    url: image-13.png
- text: a black and white photo of a small crater on mars in the style of <s0><s1>
  output:
    url: image-14.png
- text: an aerial view of a town with a river and buildings in the style of <s0><s1>
  output:
    url: image-15.png
- text: a small airplane is seen in the sky in the style of <s0><s1>
  output:
    url: image-16.png
- text: a clock and a book on a table in the style of <s0><s1>
  output:
    url: image-17.png
- text: a black and white image of a number of numbers in the style of <s0><s1>
  output:
    url: image-18.png
- text: an aerial view of the oil refinery in the 1950s in the style of <s0><s1>
  output:
    url: image-19.png
- text: a man and woman standing in a living room in the style of <s0><s1>
  output:
    url: image-20.png
- text: an aerial view of a parking lot and a road in the style of <s0><s1>
  output:
    url: image-21.png
- text: a man dressed in a costume sitting on the ground in the style of <s0><s1>
  output:
    url: image-22.png
- text: an aerial view of an airport with several planes in the style of <s0><s1>
  output:
    url: image-23.png
- text: a truck is driving through the desert in black and white in the style of <s0><s1>
  output:
    url: image-24.png
- text: a fighter jet is parked on a runway in the snow in the style of <s0><s1>
  output:
    url: image-25.png
- text: an aerial view of a tank in the desert in the style of <s0><s1>
  output:
    url: image-26.png
- text: an old black and white photo of a car in the desert in the style of <s0><s1>
  output:
    url: image-27.png
- text: an aerial view of an army vehicle and an armored personnel carrier parked
    in the desert in the style of <s0><s1>
  output:
    url: image-28.png
- text: a black and white photo of trucks in the desert in the style of <s0><s1>
  output:
    url: image-29.png
- text: an m60a1 tank is seen in the desert in the style of <s0><s1>
  output:
    url: image-30.png
- text: an m113a3 armored personnel carrier (APC) is seen in the desert in the style
    of <s0><s1>
  output:
    url: image-31.png
- text: a black and white photo of a tank in the desert in the style of <s0><s1>
  output:
    url: image-32.png
- text: a black and white photo of a truck and a car in the style of <s0><s1>
  output:
    url: image-33.png
- text: a black and white photo of a fighter jet in the style of <s0><s1>
  output:
    url: image-34.png
- text: black and white photo of boats on the shore in the style of <s0><s1>
  output:
    url: image-35.png
- text: a black and white color palette with a white and black in the style of <s0><s1>
  output:
    url: image-36.png
- text: a blue car parked in front of a house in the style of <s0><s1>
  output:
    url: image-37.png
- text: a graph paper with the coordinates of the coordinates in the style of <s0><s1>
  output:
    url: image-38.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: in the style of <s0><s1>
license: openrail++
---

# SDXL LoRA DreamBooth - lfischbe/3sipi

<Gallery />

## Model description

### These are lfischbe/3sipi LoRA adaption weights for stabilityai/stable-diffusion-xl-base-1.0.

## Download model

### Use it with UIs such as AUTOMATIC1111, Comfy UI, SD.Next, Invoke

- **LoRA**: download **[`3sipi.safetensors` here 💾](/lfischbe/3sipi/blob/main/3sipi.safetensors)**.
    - Place it on your `models/Lora` folder.
    - On AUTOMATIC1111, load the LoRA by adding `<lora:3sipi:1>` to your prompt. On ComfyUI just [load it as a regular LoRA](https://comfyanonymous.github.io/ComfyUI_examples/lora/).
- *Embeddings*: download **[`3sipi_emb.safetensors` here 💾](/lfischbe/3sipi/blob/main/3sipi_emb.safetensors)**.
    - Place it on it on your `embeddings` folder
    - Use it by adding `3sipi_emb` to your prompt. For example, `in the style of 3sipi_emb`
    (you need both the LoRA and the embeddings as they were trained together for this LoRA)
    

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

```py
from diffusers import AutoPipelineForText2Image
import torch
from huggingface_hub import hf_hub_download
from safetensors.torch import load_file
        
pipeline = AutoPipelineForText2Image.from_pretrained('stabilityai/stable-diffusion-xl-base-1.0', torch_dtype=torch.float16).to('cuda')
pipeline.load_lora_weights('lfischbe/3sipi', weight_name='pytorch_lora_weights.safetensors')
embedding_path = hf_hub_download(repo_id='lfischbe/3sipi', filename='3sipi_emb.safetensors' repo_type="model")
state_dict = load_file(embedding_path)
pipeline.load_textual_inversion(state_dict["clip_l"], token=["<s0>", "<s1>"], text_encoder=pipeline.text_encoder, tokenizer=pipeline.tokenizer)
pipeline.load_textual_inversion(state_dict["clip_g"], token=["<s0>", "<s1>"], text_encoder=pipeline.text_encoder_2, tokenizer=pipeline.tokenizer_2)
        
image = pipeline('in the style of <s0><s1>').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)

## Trigger words

To trigger image generation of trained concept(or concepts) replace each concept identifier in you prompt with the new inserted tokens:

to trigger concept `TOK` → use `<s0><s1>` in your prompt 



## Details
All [Files & versions](/lfischbe/3sipi/tree/main).

The weights were trained using [🧨 diffusers Advanced Dreambooth Training Script](https://github.com/huggingface/diffusers/blob/main/examples/advanced_diffusion_training/train_dreambooth_lora_sdxl_advanced.py).

LoRA for the text encoder was enabled. False.

Pivotal tuning was enabled: True.

Special VAE used for training: madebyollin/sdxl-vae-fp16-fix.