Dataset · hide1
Catalog of generated characters and styles with prompts
hide1/MOSS-Image-Character-Catalog
34,709 synthetic character and style images with previews and an English/Japanese prompt JSON catalog.
What's inside
The collection contains 34,395 character images and 314 style examples (34,709 generated images in total), distributed across 8 mutually exclusive categories, plus 280-pixel thumbnails. A catalog.json file is included with public names, prompts, and tags in English and Japanese. The images are synthetic and did not undergo manual filtering.
What you can do with it
- Training LoRA adapters for anime character generation and image stylization
- Testing and tuning prompts and tags for diffusion models
- Classifying images by art styles
- Languages
- English, Japanese
- Size
- 10K–100K rows
- Downloads this month
- 64.1K
- Likes
- 1
How to load it
from datasets import load_dataset
ds = load_dataset("hide1/MOSS-Image-Character-Catalog")