CoolFace
Modelpublic

GazTrab/3drenec

sourceHugging Faceopenrail++updated 2y agoView on Hugging Face
0likes45downloads
Model Card

SDXL LoRA DreamBooth - GazTrab/3drenec

<Gallery />

Model description

These are GazTrab/3drenec 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 [`3drenec.safetensors` here 💾](/GazTrab/3drenec/blob/main/3drenec.safetensors).
  • —Place it on your models/Lora folder.
  • —On AUTOMATIC1111, load the LoRA by adding <lora:3drenec:1> to your prompt. On ComfyUI just load it as a regular LoRA.
  • —Embeddings: download [`3drenec_emb.safetensors` here 💾](/GazTrab/3drenec/blob/main/3drenec_emb.safetensors).
  • —Place it on it on your embeddings folder
  • —Use it by adding 3drenec_emb to your prompt. For example, 3D illustration in the style of 3drenec_emb (you need both the LoRA and the embeddings as they were trained together for this LoRA)

Use it with the 🧨 diffusers library

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('GazTrab/3drenec', weight_name='pytorch_lora_weights.safetensors')
embedding_path = hf_hub_download(repo_id='GazTrab/3drenec', filename='3drenec_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('3D illustration in the style of <s0><s1>, depicting a sleek, modern workspace with the <s0><s1> logo intricately woven into the design of a futuristic computer interface. The desk is vibrant teal, matching the logo hue, surrounded by angular, geometric shapes that echo the <s0><s1> brand aesthetic. Above the workspace, holographic screens display interactive graphs and data analytics, subtly incorporating the <s0><s1> logo. The scene is enhanced by soft, ambient light, emphasizing cutting-edge technology and innovation.').images[0]

For more details, including weighting, merging and fusing LoRAs, check the documentation on loading LoRAs in diffusers

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.

The weights were trained using 🧨 diffusers Advanced Dreambooth Training Script.

LoRA for the text encoder was enabled. False.

Pivotal tuning was enabled: True.

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