Skip to content

Datasets

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.

Data type

For which models

Find a dataset for your goal

Answered by Gemini 4 · the chat is kept only in this tab

DatasetDSULT-Core/bluesky-298-million-Posts

Archive of 298 million Bluesky social network posts

Large-scale corpus of English microblog posts from Bluesky for text generation and classification.

  • Text generation
  • Text classification
  • English
85 49
Datasetsunlab/patch_db

PatchDB security patch dataset

A corpus of over 35,000 patches for security research and vulnerability patching.

  • Embeddings
  • Text classification
  • Summarization
  • English
85 2910K–100K rows🔒 on request
Datasetro-h/regulatory_comments_api

Comments on US opioid bills

A small English dataset with dockets from Regulations.gov and the top 10 public comments for each.

  • Text classification
  • English
75 41up to 1K rows
Datasetalibayram/doktorsitesi

Turkish medical questions and answers

Corpus of 167k Turkish medical questions and answers for text generation and classification.

  • Text generation
  • Text classification
  • Table question answering
  • Turkish
73 48100K–1M rows🔒 on request
DatasetSoftAge-AI/multi-turn_dataset

Multi-turn dialogues for SFT and DPO

A set of 400 English multi-turn dialogues for fine-tuning and alignment of language models.

  • Question answering
  • Text classification
  • Summarization
  • English
70 9up to 1K rows🔒 on request
Datasetirlab-udc/redsm5

Reddit dataset for depression detection

A Reddit text dataset for detecting signs of depression according to DSM-5.

  • Text classification
  • Text generation
  • English
66 241K–10K rows🔒 on request
DatasetGenius-Society/hoyoMusic

Genshin Impact fan music in ABC notation

Over 300,000 fragments of game piano music in ABC text format for melody generation.

  • Text generation
  • Text classification
  • English
  • Chinese
57 3110K–100K rows
DatasetHackerNoon/tech-company-news-data-dumpTop pick

HackerNoon 7M+ tech company news archive

A massive collection of articles and posts from 2022-2023 about 3,000+ top tech companies.

  • Text classification
  • Summarization
  • English
53 461M–10M rows🔒 on request