Skip to content

Dataset · jainr3

2M Stable Diffusion generations with prompts

jainr3/diffusiondb-pixelart

Large Stable Diffusion image corpus with user prompts and generation parameters.

What's inside

A large-scale multimodal dataset (1M to 10M entries) containing 2 million images generated by the Stable Diffusion model. Real English prompts and user-defined hyperparameters are attached to each image (description taken from the base DiffusionDB card).

What you can do with it

  • Training or fine-tuning text-to-image generation models
  • Creating synthetic image (deepfake) detectors
  • Training prompt retrieval and improvement models (image-to-text)
Tasks
Text → image, Image → text
Languages
English
Size
1M–10M rows
Downloads this month
297
Likes
61

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("jainr3/diffusiondb-pixelart")