Skip to content

Dataset · HappyHenAi

Images and captions without AI slop for Anima LoRA

HappyHenAi/AiDatasetV0.3-Anti-ai-slop-anima-b1

A large dataset of image-text pairs without re-compression for training diffusion models.

What's inside

A dataset ranging from 100k to 1M examples with English captions and original filenames. Images are embedded as raw bytes (PNG, JPG, WebP) without lossy re-compression, and the entire set resides in a single train split.

What you can do with it

  • Training LoRA adapters for the Anima diffusion model
  • Fine-tuning text-to-image models on images without AI visual artifacts
Languages
English
Size
100K–1M rows
Downloads this month
31.7K
Likes
0

How to load it

from datasets import load_dataset

ds = load_dataset("HappyHenAi/AiDatasetV0.3-Anti-ai-slop-anima-b1")