---
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
widget:
- text: '-'
  output:
    url: images/Beeple-Style-00001.png
- text: '-'
  output:
    url: images/Beeple-Style-00003.png
- text: '-'
  output:
    url: images/Beeple-Style-00008.png
- text: '-'
  output:
    url: images/Beeple-Style-00010.png
- text: '-'
  output:
    url: images/Beeple-Style-00016.png
- text: '-'
  output:
    url: images/Beeple-Style-00020.png
- text: '-'
  output:
    url: images/Beeple-Style-00021.png
- text: '-'
  output:
    url: images/Beeple-Style-00027.png
- text: '-'
  output:
    url: images/Beeple-Style-00031.png
- text: '-'
  output:
    url: images/Beeple-Style-00032.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: beeple style, digital art, meticulous
license: artistic-2.0
---
# Beeple Style

<Gallery />

## Model description 

Beeple Style
In respect for Beeple day and the infamous $69 trillion jporg here is a LoRA trained on 32 recent Beeple daily works. It does a good job of making large subjects with small people and does well at recreating his Topaz Studio painterly post FX look. 32 Beeple images were captioned with GPTV and trained for 40 epochs at 1280 x 1600 with a batch size of 4 and gradient accumulation of 2 using the prodigy optimizer, after some quick testing i chose epoch 38 to share. Demo images were generated using the SDXL 1.0 base model along with the tempest model for the refiner at 20 more steps.

Now you are Beeple, go make a trillion dollar jpg! Also if you are actually Beeple hit me up!

## Trigger words

You should use `beeple style` to trigger the image generation.

You should use `digital art` to trigger the image generation.

You should use `meticulous` to trigger the image generation.


## Download model

Weights for this model are available in Safetensors format.

[Download](/BlaireSilver13/Beeple_Style/tree/main) them in the Files & versions tab.