Crossref API vs OpenAbstracts: metadata lookup vs paper search
Published September 26, 2026
Crossref and OpenAbstracts solve different problems, and the best setups use both. Crossref is the registry behind DOIs; OpenAbstracts is a search API for finding and quoting papers. Details are as of September 2026 from Crossref's public documentation.
The short version
- Use Crossref as the authoritative source for DOI metadata: titles, authors, funding, licences, ORCID and ROR identifiers, post-publication updates and, where deposited, abstracts. It is free, needs no sign-up, and its metadata may be used for almost any purpose.
- Use OpenAbstracts when you want to search by topic or phrase — including inside PDF full text — with relevance-ranked results, highlighted snippets, journal-quartile filtering and PDF links, in a single call.
What Crossref provides
According to Crossref's documentation, its REST API exposes scholarly metadata that members deposit, including bibliographic data, funding and licence information, ORCID and ROR IDs, and abstracts. No sign-up is required, and it supports searching, faceting, filtering and sampling across works, journals, members, funders and more. A "polite pool" is available if you identify yourself with a mailto parameter, and a paid Metadata Plus service exists separately.
Crossref's coverage of who published what, and its identifiers is unmatched. Its search, however, is a lookup over deposited metadata — it isn't a full-text or quality-filtered discovery engine, and abstract availability depends on what each publisher deposited.
Side by side
| Crossref REST API | OpenAbstracts | |
|---|---|---|
| Purpose | Authoritative DOI and publication metadata | Search and discovery over papers from registered journals |
| Cost | Free, no sign-up (paid Metadata Plus separately) | 100 free searches/month, then $1.50 per 1,000 |
| Search | Filter and query over deposited metadata | Relevance-ranked keyword search and full-text PDF search |
| Snippets | No | Highlighted passages from PDF text |
| Quartile/quality filter | No | quartile, category, yearFrom/yearTo |
| AI assistants | REST API | MCP server with OAuth Connect |
Using them together
A robust citation pipeline often looks like this:
- Discover with OpenAbstracts: search by topic, filter by quartile and year, read abstracts and snippets.
- Confirm with Crossref: resolve the DOI to get the canonical, publisher-deposited metadata and licence.
- Present with the source link the API returned.
# 1. discover
curl "https://openabstracts.com/api/references-search?q=circular+economy+supply+chains&quartile=Q1&limit=3" \
-H "x-api-key: isk_live_YOUR_KEY"
# 2. confirm a DOI (free, no key)
curl "https://api.crossref.org/works/10.12775/EQUIL.2009.011"
If your concern is fabricated references, see citation verification for AI writing tools.
Bottom line
Crossref answers "what is this DOI?" better than anything; OpenAbstracts answers "which papers should I read or cite for this topic?". Start with the quickstart.
Sources
Crossref REST API documentation. OpenAbstracts pricing: docs.