README.md

---
tags:
- stable-diffusion-xl
- stable-diffusion-xl-diffusers
- diffusers-training
- text-to-image
- diffusers
- lora
- template:sd-lora
widget:

        - text: 'a photo of living room with <s0><s1> and <s2><s3>'
          output:
            url:
                "image_0.png"
        
        - text: 'a photo of living room with <s0><s1> and <s2><s3>'
          output:
            url:
                "image_1.png"
        
        - text: 'a photo of living room with <s0><s1> and <s2><s3>'
          output:
            url:
                "image_2.png"
        
        - text: 'a photo of living room with <s0><s1> and <s2><s3>'
          output:
            url:
                "image_3.png"
        
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: a photo of living room with <s0><s1> and <s2><s3> in scandinavic style
license: openrail++
---

# SDXL LoRA DreamBooth - prampampam/big-sdxl-lora

<Gallery />

## Model description

### These are prampampam/big-sdxl-lora 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 **[`big-sdxl-lora.safetensors` here 💾](/prampampam/big-sdxl-lora/blob/main/big-sdxl-lora.safetensors)**.
    - Place it on your `models/Lora` folder.
    - On AUTOMATIC1111, load the LoRA by adding `<lora:big-sdxl-lora:1>` to your prompt. On ComfyUI just [load it as a regular LoRA](https://comfyanonymous.github.io/ComfyUI_examples/lora/).
- *Embeddings*: download **[`big-sdxl-lora_emb.safetensors` here 💾](/prampampam/big-sdxl-lora/blob/main/big-sdxl-lora_emb.safetensors)**.
    - Place it on it on your `embeddings` folder
    - Use it by adding `big-sdxl-lora_emb` to your prompt. For example, `a photo of living room with big-sdxl-lora_emb and  in scandinavic style`
    (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('prampampam/big-sdxl-lora', weight_name='pytorch_lora_weights.safetensors')
embedding_path = hf_hub_download(repo_id='prampampam/big-sdxl-lora', filename='big-sdxl-lora_emb.safetensors', repo_type="model")
state_dict = load_file(embedding_path)
pipeline.load_textual_inversion(state_dict["clip_l"], token=["<s0>", "<s1>", "<s2>", "<s3>"], text_encoder=pipeline.text_encoder, tokenizer=pipeline.tokenizer)
pipeline.load_textual_inversion(state_dict["clip_g"], token=["<s0>", "<s1>", "<s2>", "<s3>"], text_encoder=pipeline.text_encoder_2, tokenizer=pipeline.tokenizer_2)
        
image = pipeline('a photo of living room with <s0><s1> and <s2><s3>').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 `TOK49986` → use `<s0><s1>` in your prompt 


to trigger concept `TOK52817` → use `<s2><s3>` in your prompt 


to trigger concept `TOK98678` → use `<s4><s5>` in your prompt 


to trigger concept `TOK113071` → use `<s6><s7>` in your prompt 


to trigger concept `TOK120149` → use `<s8><s9>` in your prompt 


to trigger concept `TOK123556` → use `<s10><s11>` in your prompt 


to trigger concept `TOK135534` → use `<s12><s13>` in your prompt 


to trigger concept `TOK189670` → use `<s14><s15>` in your prompt 


to trigger concept `TOK190562` → use `<s16><s17>` in your prompt 


to trigger concept `TOK193714` → use `<s18><s19>` in your prompt 


to trigger concept `TOK241048` → use `<s20><s21>` in your prompt 


to trigger concept `TOK250041` → use `<s22><s23>` in your prompt 


to trigger concept `TOK268959` → use `<s24><s25>` in your prompt 


to trigger concept `TOK273000` → use `<s26><s27>` in your prompt 


to trigger concept `TOK277813` → use `<s28><s29>` in your prompt 


to trigger concept `TOK279746` → use `<s30><s31>` in your prompt 


to trigger concept `TOK295935` → use `<s32><s33>` in your prompt 


to trigger concept `TOK305227` → use `<s34><s35>` in your prompt 


to trigger concept `TOK309606` → use `<s36><s37>` in your prompt 


to trigger concept `TOK321136` → use `<s38><s39>` in your prompt 


to trigger concept `TOK329781` → use `<s40><s41>` in your prompt 


to trigger concept `TOK329782` → use `<s42><s43>` in your prompt 


to trigger concept `TOK352259` → use `<s44><s45>` in your prompt 


to trigger concept `TOK355638` → use `<s46><s47>` in your prompt 


to trigger concept `TOK357399` → use `<s48><s49>` in your prompt 


to trigger concept `TOK368921` → use `<s50><s51>` in your prompt 


to trigger concept `TOK377236` → use `<s52><s53>` in your prompt 


to trigger concept `TOK382924` → use `<s54><s55>` in your prompt 


to trigger concept `TOK405550` → use `<s56><s57>` in your prompt 


to trigger concept `TOK429026` → use `<s58><s59>` in your prompt 


to trigger concept `TOK434773` → use `<s60><s61>` in your prompt 


to trigger concept `TOK435237` → use `<s62><s63>` in your prompt 


to trigger concept `TOK437871` → use `<s64><s65>` in your prompt 


to trigger concept `TOK439185` → use `<s66><s67>` in your prompt 


to trigger concept `TOK439384` → use `<s68><s69>` in your prompt 


to trigger concept `TOK443869` → use `<s70><s71>` in your prompt 


to trigger concept `TOK444874` → use `<s72><s73>` in your prompt 


to trigger concept `TOK453903` → use `<s74><s75>` in your prompt 


to trigger concept `TOK465250` → use `<s76><s77>` in your prompt 


to trigger concept `TOK482466` → use `<s78><s79>` in your prompt 


to trigger concept `TOK503940` → use `<s80><s81>` in your prompt 


to trigger concept `TOK524768` → use `<s82><s83>` in your prompt 


to trigger concept `TOK536204` → use `<s84><s85>` in your prompt 


to trigger concept `TOK537466` → use `<s86><s87>` in your prompt 


to trigger concept `TOK547839` → use `<s88><s89>` in your prompt 


to trigger concept `TOK548004` → use `<s90><s91>` in your prompt 


to trigger concept `TOK549578` → use `<s92><s93>` in your prompt 


to trigger concept `TOK557244` → use `<s94><s95>` in your prompt 


to trigger concept `TOK561859` → use `<s96><s97>` in your prompt 


to trigger concept `TOK569103` → use `<s98><s99>` in your prompt 


to trigger concept `TOK569104` → use `<s100><s101>` in your prompt 


to trigger concept `TOK570087` → use `<s102><s103>` in your prompt 


to trigger concept `TOK577885` → use `<s104><s105>` in your prompt 


to trigger concept `TOK578163` → use `<s106><s107>` in your prompt 


to trigger concept `TOK586036` → use `<s108><s109>` in your prompt 


to trigger concept `TOK590747` → use `<s110><s111>` in your prompt 


to trigger concept `TOK592708` → use `<s112><s113>` in your prompt 


to trigger concept `TOK597750` → use `<s114><s115>` in your prompt 


to trigger concept `TOK603453` → use `<s116><s117>` in your prompt 


to trigger concept `TOK618877` → use `<s118><s119>` in your prompt 


to trigger concept `TOK619268` → use `<s120><s121>` in your prompt 


to trigger concept `TOK629015` → use `<s122><s123>` in your prompt 


to trigger concept `TOK631826` → use `<s124><s125>` in your prompt 


to trigger concept `TOK639063` → use `<s126><s127>` in your prompt 


to trigger concept `TOK639380` → use `<s128><s129>` in your prompt 


to trigger concept `TOK648242` → use `<s130><s131>` in your prompt 


to trigger concept `TOK651623` → use `<s132><s133>` in your prompt 


to trigger concept `TOK653420` → use `<s134><s135>` in your prompt 


to trigger concept `TOK654567` → use `<s136><s137>` in your prompt 


to trigger concept `TOK657274` → use `<s138><s139>` in your prompt 


to trigger concept `TOK658513` → use `<s140><s141>` in your prompt 


to trigger concept `TOK662375` → use `<s142><s143>` in your prompt 


to trigger concept `TOK664555` → use `<s144><s145>` in your prompt 


to trigger concept `TOK674941` → use `<s146><s147>` in your prompt 


to trigger concept `TOK679104` → use `<s148><s149>` in your prompt 



## Details
All [Files & versions](/prampampam/big-sdxl-lora/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.