33

I wonder whether it is possible to bulk download all papers stored in Sci-Hub.


I am aware of the questions:

but they focus on arXiv.

2

3 Answers 3

30

Sci-Hub is a paywall-bypassing website that uses "shared" user credentials to provide PDF or HTML scientific papers. The website itself doesn't store any papers. (There are interesting comments on your same question on another site.)

But LibGen (via wayback machine) is said to archive each PDF retrieved by Sci-Hub.

http://gen.lib.rus.ec has a downloads page - (click the tab to show more options)

enter image description here

10
  • 5
    bonus points for .torrent open data
    – albert
    Commented Feb 14, 2016 at 15:33
  • 4
    yeah - someone can add it to the community wiki opendata.stackexchange.com/a/4088/1511
    – philshem
    Commented Feb 14, 2016 at 16:30
  • 3
    done. although i wasn't sure what to add for the torrent size.....the sum of all the torrents in the folder?
    – albert
    Commented Feb 14, 2016 at 17:09
  • 1
    comment placeholder for @myrix: A slight correction/addition to philshem's answer. Not a comment due to not having enough reputation. If you need specifically scientific papers, you should use torrents at the following link: libgen.io/scimag/repository_torrent_notforall Collection of data accessible through torrents at libgen.io/repository_torrent, mentioned by philshem, is focused on books, including textbooks.
    – albert
    Commented Mar 24, 2016 at 6:26
  • 2
    @albert Thanks, libgen.io/scimag/repository_torrent_notforall is indeed the way to go for scientific papers. Around 53 million papers, ~18 TB. Commented May 30, 2016 at 14:07
5

The Reddit thread Can I download all libgen books? describes how to download the entire Library Genesis database (~10TB).

2
0

Another resource which is a legal alternative to sci-hub.tw is Unpaywall.

We harvest content directly from over 50,000 journals and open-access repositories from all over the world. We also use great open data from PubMed Central, the DOAJ, Crossref (particulary their license info), and DataCite.


  • Rest API: takes a DOI and returns JSON data with URLs to paper PDF

Example: https://api.unpaywall.org/v2/10.1038/nature12373?email=YOUR_EMAIL


To complete the download of the full dataset, please fill out this form, which helps us report usage back to our funders, and we'll immediately provide you with a download link.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.