scholarly-metadata
Connected Papers
A tool that builds a visual graph of papers related to a given work by similarity, drawn from a corpus of tens of millions of papers. You use it by opening the graph view for a paper on its website. There is no open lookup interface, so it serves as a deep link, reach for it to discover adjacent literature around a paper you already know.
No programmatic check, opens the archive’s own search.
Why it’s useful & how it works
Connected Papers is built for the early stage of a literature search, when you know one relevant paper and want to discover what else is nearby. It builds a visual graph where papers are positioned by how strongly they co-cite each other, so clusters of closely related work become immediately visible. You use it by opening the graph for a seed paper on connectedpapers.com. There is no public open lookup interface, so Echo links you to the graph page for the paper you have in hand, letting you explore related literature visually from there.
What’s inside
Draws on the Semantic Scholar corpus of hundreds of millions of papers across all scientific fields. Each graph explores around 50,000 candidate papers to surface the 40 to 50 most relevant to the chosen seed work.
API access
official Python/JS clients (key, ~5 builds/min)
An API key is required (usually free); see the endpoints above for where to get one.
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 1.1 s
- Through a datacenter proxy
- Responded HTTP 200 816 ms
Reachability measured 2026-08-22.
Access
Catalogue link only, open the archive to search it yourself.