Downloads · 30 days
38
1% of all-time downloads
renderartist/ROYGBIVFlux
ROYGBIVFlux is a text-to-image model from renderartist. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as creativeml-openrail-m.
Downloads · 30 days
38
1% of all-time downloads
All-time downloads
7.2K
Public
Repo size
485 MB
Likes
25
Public
Click a slice to open those files.
.safetensors460 MB · 95%
From the Hugging Face model README
ROYGBIV Flux blends elements inspired by risograph and textured painting techniques to create a unique style that provides images a striking visual appeal with a storybook feel.
ROYGBIV v2 was trained on 149 AI-generated images, 2 batch, 1 repeat, 24 DIM / 32 ALPHA -- 2,600 training steps.
I'm sharing two versions of this LoRA:
v1 - Requires more prompting to achieve the effect but has more general flexibility in terms of composition.
v2 - Works with less prompting but can be a little more rigid in terms of flexibility.
Still trying to work out the best settings for training this style but it's still very usable and provides some great results with a little forethought and creative prompting. Check out the prompts to get an idea of certain phrases that achieve the look you're trying to focus on.
FOR THE BEST RESULTS USE EULER SAMPLER
Trigger keyword: 'r0ygb1v' 'digital illustration' 'textured'
Recommended strengths: 0.7 - 0.9
Trained in Kohya, tested in ComfyUI.
You should use r0ygb1v to trigger the image generation.
You should use digital illustration to trigger the image generation.
You should use textured to trigger the image generation.
Weights for this model are available in Safetensors format.
Download them in the Files & versions tab.