Dataset · moonworks
Moonworks Lunara artistic variations
moonworks/lunara-aesthetic-image-variations
Original art and its contextual variations for image-to-image and styling tasks.
What's inside
A set of original images and artwork from Moonworks and contextual variations generated by the Moonworks Lunara model. The model has under 10B parameters and a novel diffusion mixture architecture. Volume ranges from 1K to 10K examples, modalities are image and text; part 1 is intended for studying and evaluating regional and region-agnostic artistic styles.
What you can do with it
- Training LoRA for image styling
- Testing image-to-image models on art variations
- Evaluating artistic style transfer
- Tasks
- Image → image, Text → image
- Size
- 1K–10K rows
- Downloads this month
- 629
- Likes
- 72
How to load it
from datasets import load_dataset
ds = load_dataset("moonworks/lunara-aesthetic-image-variations")