metadata

license: other
license_name: bespoke-lora-trained-license
license_link: >-
  https://multimodal.art/civitai-licenses?allowNoCredit=True&allowCommercialUse=Sell&allowDerivatives=True&allowDifferentLicense=True
tags:
  - text-to-image
  - stable-diffusion
  - lora
  - diffusers
  - template:sd-lora
  - vibrant colors
  - style
  - hires fix
  - detail
  - quality up
  - sharpness
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: null
widget:
  - text: ' '
    output:
      url: 4449094.jpeg
  - text: ' '
    output:
      url: 4449095.jpeg
  - text: ' '
    output:
      url: 4449092.jpeg
  - text: ' '
    output:
      url: 4449100.jpeg
  - text: >-
      photograph, cinematic color grading, 1girl, woman, beautiful, Industrial
      loft with skyline view in a empty fantasy topia outside of time,
      masterpiece
    output:
      url: 4449312.jpeg

Envy Better Hires Fix XL 01

Prompt
photograph, cinematic color grading, 1girl, woman, beautiful, Industrial loft with skyline view in a empty fantasy topia outside of time, masterpiece

Model description

Adds contrast, detail, and quality improvements to hires fix. Improves hands a bit. Gives more realistic skin textures on photorealistic models.

Important generation notes:

In the example images, this LoRA was only enabled during the Hires Fix step. I've had very good results enabling it for both passes (in which case it makes generations more dramatic and vibrant), but it does alter composition somewhat, which makes direct comparison harder.

If you want to run this LoRA during your hires fix only, you can download this a1111 extension here:

https://github.com/cheald/sd-webui-loractl

...and add the LoRA to your positive prompt like this: <lora:EnvyBetterHiresFixXL01:0:hr=1>

(Or just add it like a normal LoRA to use it during both passes.)

Download model

Weights for this model are available in Safetensors format.

Download them in the Files & versions tab.

Use it with the 🧨 diffusers library

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('e-n-v-y/envy-better-hires-fix-xl-01', weight_name='EnvyBetterHiresFixXL01.safetensors')
image = pipeline('photograph, cinematic color grading, 1girl, woman, beautiful, Industrial loft with skyline view in a empty fantasy topia outside of time, masterpiece').images[0]

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