README.md

---
tags:
- text-to-image
- stable-diffusion
- lora
- diffusers
- template:sd-lora
widget:
- text: '-'
  output:
    url: images/kodachrome_00106_.png
- text: '-'
  output:
    url: images/kodachrome_00077_.png
- text: '-'
  output:
    url: images/kodachrome_00075_.png
- text: '-'
  output:
    url: images/kodachrome_00072_.png
- text: '-'
  output:
    url: images/kodachrome_00036_.png
- text: '-'
  output:
    url: images/kodachrome_00027_.png
- text: '-'
  output:
    url: images/kodachrome_00030_.png
- text: '-'
  output:
    url: images/kodachrome_00017_.png
- text: '-'
  output:
    url: images/kodachrome_00016_.png
- text: '-'
  output:
    url: images/kodachrome_00012_-1.png
- text: '-'
  output:
    url: images/kodachrome_00003_.png
- text: '-'
  output:
    url: images/cryptocollege-_00142_.png
- text: '-'
  output:
    url: images/cryptocollege-_00127_.png
base_model: stabilityai/stable-diffusion-xl-base-1.0
instance_prompt: null
license: artistic-2.0
---
# Kodachrome

<Gallery />

## Model description 

This LoRA attempts to capture the look, color profile, softness and grain of Kodachrome film photographs. It also captures vintage aesthetics in clothing, cars and overall style. Trained from 88 images scraped from the kodachrome group on flickr and various other photo sharing websites. Trained for 50 epochs with a batch size of 8 and a gradient accumulation of 8 using the prodigy optimizer and a network strength of 128. While no trigger words are needed, using &quot;Shot on Kodachrome film&quot; and &quot;film grain&quot; can help push the look.


## Download model

Weights for this model are available in Safetensors format.

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