Skip to content

Dataset · poloclub

2M Stable Diffusion images with prompts

poloclub/diffusiondb

2M Stable Diffusion images with prompts and hyperparameters from real users.

What's inside

DiffusionDB is the first large-scale text-to-image prompt dataset. It contains 2M images generated by Stable Diffusion using prompts and hyperparameters specified by real users. Prompts are in English; metadata includes generation parameters.

What you can do with it

  • Training and evaluating text-to-image models
  • Detecting generated images and deepfakes
  • Studying the relationship between prompts and generative models
Tasks
Text → image, Image → text
Languages
English
Size
over 1T rows
Downloads this month
29.2K
Likes
678

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("poloclub/diffusiondb")