---
tags:
- stable-diffusion-xl
- stable-diffusion-xl-diffusers
- text-to-image
- diffusers
- lora
- template:sd-lora
widget:
- text: a drawing of a woman's head with many different colored lines in the style
    of <s0><s1>. The image is a colorful drawing of a person's face. The face is drawn
    in a cartoon-like style with bright colors and intricate details. The person's
    eyes are large and expressive, with a red nose and red lips. They are wearing
    a blue shirt with a pink collar. The background of the drawing is white, and it
    is filled with various shapes and patterns, including triangles, squares, and
    rectangles in various colors such as green, yellow, red, and blue. The shapes
    are arranged in a way that creates a sense of movement and energy, with some overlapping
    and some standing out. The overall effect is one of chaos and confusion.
  output:
    url: image-0.png
- text: a painting of a woman's face with blue eyes in the style of <s0><s1>. painful
    lightning shoots between the eyes.
  output:
    url: image-1.png
- text: a drawing of a person laying in bed with their eyes open in the style of <s0><s1>
  output:
    url: image-2.png
- text: a drawing of four people with different colored heads in the style of <s0><s1>
  output:
    url: image-3.png
- text: a drawing of a woman with a flower in her hair in the style of <s0><s1>
  output:
    url: image-4.png
- text: a painting with a woman's face and eye in the style of <s0><s1>
  output:
    url: image-5.png
- text: a drawing of a man in a room with a green and yellow wall in the style of
    <s0><s1>
  output:
    url: image-6.png
- text: a painting of a face with stars on it in the style of <s0><s1>
  output:
    url: image-7.png
- text: a painting of a wave with a white circle around it in the style of <s0><s1>
  output:
    url: image-8.png
- text: a drawing of two birds on a post with flowers in the style of <s0><s1>
  output:
    url: image-9.png
- text: a drawing of a classroom with a teacher and students in the style of <s0><s1>
  output:
    url: image-10.png
- text: a painting of a bridge with a green tunnel in the style of <s0><s1>
  output:
    url: image-11.png
- text: a painting of a woman with blood on her face in the style of <s0><s1>
  output:
    url: image-12.png
- text: a drawing of a woman with a red flower in her head in the style of <s0><s1>
  output:
    url: image-13.png
- text: a black and white drawing of waves on a white background in the style of <s0><s1>
  output:
    url: image-14.png
- text: a drawing of a man and woman with a broken face in the style of <s0><s1>
  output:
    url: image-15.png
- text: a drawing of a colorful pattern of dots in the style of <s0><s1>
  output:
    url: image-16.png
- text: an abstract painting of a woman's face with blue eyes in the style of <s0><s1>
  output:
    url: image-17.png
- text: a painting of five women with different faces in the style of <s0><s1>
  output:
    url: image-18.png
- text: a painting of a woman with a sunburst behind her in the style of <s0><s1>
  output:
    url: image-19.png
- text: a painting of a woman with a green face and yellow eyes in the style of <s0><s1>
  output:
    url: image-20.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: in the style of <s0><s1>
license: openrail++
---

# SDXL LoRA DreamBooth - lfischbe/m1gr

<Gallery />

## Model description

### These are lfischbe/m1gr 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 **[`m1gr.safetensors` here 💾](/lfischbe/m1gr/blob/main/m1gr.safetensors)**.
    - Place it on your `models/Lora` folder.
    - On AUTOMATIC1111, load the LoRA by adding `<lora:m1gr:1>` to your prompt. On ComfyUI just [load it as a regular LoRA](https://comfyanonymous.github.io/ComfyUI_examples/lora/).
- *Embeddings*: download **[`m1gr_emb.safetensors` here 💾](/lfischbe/m1gr/blob/main/m1gr_emb.safetensors)**.
    - Place it on it on your `embeddings` folder
    - Use it by adding `m1gr_emb` to your prompt. For example, `in the style of m1gr_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/m1gr', weight_name='pytorch_lora_weights.safetensors')
embedding_path = hf_hub_download(repo_id='lfischbe/m1gr', filename='m1gr_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/m1gr/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.