400k Pexels Videos with Captions and Metadata
400,476 videos with previews, view counts, explicit flags, and captions for video-text models.
- Image → text
- Text → image
- Text → video
- English
Hugging Face datasets with plain-language descriptions: what's inside and what models to train on them. 4,788 in the catalog so far, with new ones added every hour.
Answered by Gemini 4 · the chat is kept only in this tab
400,476 videos with previews, view counts, explicit flags, and captions for video-text models.
Image and caption pairs with identical words in different orders to test vision-language models.
45K text-to-image and 46K text+image-to-image examples for training multimodal models.
A large dataset of game sprites on a transparent background with English captions for text-to-image training.
Large collection of 25 million image URLs from Common Crawl with license annotations for image generation.
1.4M English text-image pairs with high artistic style consistency.
A small dataset of Pokemon images (under 1,000 items) re-captioned with GPT-4 Turbo for image generation.
A collection of 4,172 scanned engravings and drawings from old books with text captions for image generation and analysis.
Recompressed Danbooru 2023 version: cleaned images up to 4 MP for training generative models.
Fresh dump of tags and metadata for millions of images from the Danbooru imageboard in parquet format.
Large-scale synthetic dataset for image restoration and super-resolution tasks in games.
A large Tencent dataset with long video clips and detailed text descriptions for training video generation.
8,000 generated images with prompts and evaluations from eight text-to-image models to compare their artistic qualities.
Small set of Pokemon images with bilingual captions for training text-to-image models.
13k images from OpenAI 4o with over 200k real human ratings for training preference models.
A dataset of 308K synthetic pixel art images paired with text descriptions.
Over 1.5 million ratings from real people for images generated from English text prompts.
Detailed synthetic descriptions of 16.9 million images for training and evaluating vision-language models.
Synthetic dataset with images generated by GPT-4o to improve text-to-image models.
A collection of images and masks for evaluating analog media damage segmentation models.
Original art and its contextual variations for image-to-image and styling tasks.
Huge archive of 23.1M original Midjourney images and their English text descriptions.
2,000 image-prompt pairs for research on prompt grounding, style, and aesthetics in text-to-image.
A massive corpus of 57.3M short and long English captions for 10M Flickr images from various VLM models.