What’s Out There
← Back to swiping
Language & textFreeNonprofit

Project Gutenberg

Project Gutenberg Literary Archive Foundation

Tens of thousands of public-domain books, hand-proofed by volunteers since 1971.

A library of free ebooks, mostly older literature whose U.S. copyright has expired, with a machine-readable catalog of titles, authors, subjects, and languages. It is a standard source of clean book-length text for computational literary studies, readability research, and language models.

Access

Cost
FreeFree; bulk downloading should go through mirrors or rsync rather than the main site.
License
Public domainPublic domain in the U.S. for most texts; the Project Gutenberg License governs use of its trademark and headers
Sign-up
No sign-up
How you get it
Bulk downloadPlain text, EPUB, HTML, RDF catalog

What's in it

One row is
One record per ebook, with plain-text and other file formats
Covers
Books in dozens of languages, mostly English
Years
Works mostly published before about 1930; project began 1971
Updated
Daily
Size
Over 70,000 ebooks
titleauthorauthor birth and death yearslanguagesubjectsLibrary of Congress classdownload count

Quality

4 of 5

Texts are proofread and stable, but edition information is often thin and the collection reflects volunteers' choices, not a sample of what was published.

  • Source edition and original publication date are often missing
  • Copyright status outside the U.S. varies by work
  • Collection skews toward canonical English and European literature

Connects to

Questions it could answer

  • How did average sentence length change in English novels from 1800 to 1920?
  • Which authors' books are downloaded most, relative to how many they wrote?

Similar data sets

  • Open Library Data Dumps

    A wiki-style catalog of tens of millions of book editions, dumped monthly for anyone to download.

    Film, TV, & booksInternet Archive (Open Library)Free

  • Common Corpus

    2.27 trillion tokens of text that is either uncopyrighted or openly licensed, with per-document provenance.

    Language & textPleiasFree

  • LibriSpeech ASR Corpus

    1,000 hours of volunteers reading public-domain audiobooks aloud, the speech recognition yardstick.

    Audio & musicOpenSLR (Vassil Panayotov, Daniel Povey, and colleagues)Free

  • Mozilla Common Voice

    Volunteers reading sentences aloud in 100+ languages, donated to the public domain.

    Audio & musicMozilla FoundationFree

  • The Pile

    The 825 GB open training set that taught the field how LLM data is assembled, and got sued over it.

    Language & textEleutherAIFree

  • Archive Team Twitter Stream Grab

    Twelve years of Twitter's free 1% sample stream, saved month by month until the tap closed.

    Social mediaArchive Team (hosted by the Internet Archive)Free