README.md

---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: >-
    botanical drawing, roses and onions, colored ink drawing, pale colours, high
    detail,intricately detailed, hyperdetailed volumetric lighting,
  output:
    url: images/ComfyUI_00007_.png
- text: >-
    botanical drawing, roses and onions, colored ink drawing, pale colours, high
    detail,intricately detailed, hyperdetailed volumetric lighting,
  output:
    url: images/ComfyUI_00006_.png
- text: a woman walking a dog
  output:
    url: images/ComfyUI_00011_.png
- text: a woman walking a dog
  output:
    url: images/ComfyUI_00008_.png
- text: >-
    a man eating lunch in a fancy resturant. the resturant is decorated in muted
    tones with gold trimming. he is sitting by a window that looks out at a busy
    city street
  output:
    url: images/ComfyUI_00013_.png
- text: >-
    a man eating lunch in a fancy restaurant. the restaurant is decorated in
    muted tones with gold trimming. he is sitting by a window that looks out at
    a busy city street
  output:
    url: images/ComfyUI_00012_.png
- text: a man eating lunch in a fancy restaurant
  output:
    url: images/ComfyUI_00017_.png
- text: a man eating lunch in a fancy restaurant
  output:
    url: images/ComfyUI_00016_.png
- text: a man eating lunch
  output:
    url: images/ComfyUI_00022_.png
- text: a man eating lunch
  output:
    url: images/ComfyUI_00021_.png
- text: a cat sitting on a mat
  output:
    url: images/ComfyUI_00026_.png
- text: a cat sitting on a mat
  output:
    url: images/ComfyUI_00024_.png
- text: a cat
  output:
    url: images/ComfyUI_00029_.png
- text: a cat
  output:
    url: images/ComfyUI_00028_.png
base_model: stabilityai/stable-diffusion-3.5-large
instance_prompt: wnoene
license: apache-2.0
---
# Cubic Abstract 2

<Gallery />

## Model description 

This is a style lora, designed to give an abstract feel to your Stable Diffusion 3.5 Large generations. The trigger word is wnoene.

This lora, and the Cubic Abstract 1 lora are trained from the same data set. they are very similar - however there are differences in what they do, sometimes subtle, sometimes very obvious - depending on what you prompt for.

IMPORTANT NOTE: Because this is a style lora, short prompts without a lot of details, that mostly focus on the subject, work much better than longer, more detailed prompts.

if using the latest version of comfyUI, use this

&lt;lora:wnoene-lora.safetensors:1.0&gt;

rather than the trigger word at the start of your prompt

## Trigger words

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


## Download model

Weights for this model are available in Safetensors format.

[Download](/crystalwizard/cubic-abstract-2/tree/main) them in the Files & versions tab.