scholarly-metadata

Unpaywall

A database that maps DOIs to legal open-access copies of papers, tracking OA status for around 150 million DOIs. Give it a DOI and it reports whether a free, legitimate full-text copy exists and where, via a simple free API. The standard way to answer 'can I read this paywalled paper for free somewhere legitimate?'

API

Search this archive

Why it’s useful & how it works

Reach for Unpaywall when you want to know whether a specific paper is available legally for free somewhere on the web. Give it any DOI and it searches more than 50,000 open-access sources, including repositories, preprint servers, and publisher OA pages, to tell you whether a free full-text copy exists and where. The lookup is instant via a free API that requires no registration, just your email address as a parameter. It is the fastest way to answer the question: is there a legitimate free version of this paywalled article?

What’s inside

Tracks open-access status for a number of DOIs comparable to the Crossref registry, currently in the range of roughly 150 to 180 million DOIs, with around a quarter of Crossref DOIs finding at least one free legal copy.

API access

https://api.unpaywall.org/v2/ <doi>?email=<email>

What we measured

Our own probes, not the archive’s own claims. Re-run periodically; every reading below is dated.

Reachability

Direct request: Responded HTTP 200 534 ms
Through a datacenter proxy: Responded HTTP 200 763 ms
API, direct: Responded HTTP 200 476 ms
API, through a proxy: Responded HTTP 200 657 ms

Did a real search return anything?

Returned results · median 473 ms · stable across runs

DOI->OA lookup works (the real use); the /v2/search endpoint is unreliable. Integrate as per-DOI lookup.

Sample query: Measured measurement

Benchmark coverage

Found 0 of 8 test items (0%) in the Academic papers cluster. Median response 150 ms.

By how well known the item is: 0/3 well-known, 0/3 mid, 0/2 obscure .

Reachability measured 2026-08-22. Search checked 2026-08-22. Benchmark run 2026-08-22.