The OpenAlex API here searches OpenAlex, the open index of scholarly work, and returns one typed row per paper — title, authors, publication year, citation count, DOI, work type, host venue, open-access status and URL, and the abstract, reconstructed from OpenAlex's inverted index back into readable text.
This is the practical alternative to Google Scholar, which has no public API and actively blocks automated access. OpenAlex is keyless and well-behaved, so a light query needs no help — but a large literature pull benefits from rotating exits and, more importantly, from getting the abstract rebuilt and every field flattened into one row instead of navigating OpenAlex's nested JSON. You can filter to work published from a given year onward.
$0.0004 per delivered work, up to 200 works per run. Nothing delivered means nothing charged.
You pass a query that runs across title, abstract and metadata, and optionally a year_from to restrict to recent work. Each matching work becomes one flat row.
abstract string so you can read or embed it.is_open_access and oa_url tell you when a free full text exists and where, separate from the DOI.query is required and searches title, abstract and metadata together. year_from optionally drops anything published before that year. max_results caps delivery at 200 works, and you are billed only for works returned.
One row per work. authors is an array of names; cited_by_count is the citation signal; doi, venue and type place the work; is_open_access with oa_url point to free full text when it exists; abstract is the rebuilt text; and url is the OpenAlex work page. Fields a work lacks come back null rather than guessed.
$0.0004 per delivered work ($0.40 per 1,000). Nothing delivered means nothing charged, and the $2 monthly free credit covers roughly 5,000 works here. Volume tiers take up to 30% off.
$0.0004 per delivered paper — $0.40 per 1,000 — no subscription. The $2/month free allowance is about 5,000 papers with no card.
Per-delivered-row billing means a tightly scoped query plus year_from that returns 60 works costs 60 rows. It shares a key and schema with the openFDA API if your research spans literature and safety data.
Two comparisons matter here. Against OpenAlex directly: it is free and keyless, so query it yourself for small jobs — the collector's value is the reconstructed abstract, the flat schema and rotation at scale, not access. Against Google Scholar: Scholar is what people reach for, but it has no API and blocks scrapers hard, so there is no clean, permitted way to pull it in bulk.
OpenAlex covers the same territory — hundreds of millions of works with citations and venues — through an interface built to be queried. Using it as the Scholar alternative is the honest, sustainable choice, and this collector makes its output drop straight into a table.
The DIY version is a paged OpenAlex query plus the inverted-index-to-text step for every abstract and a flattener for authors, venue and open-access fields. It is a modest amount of code you then maintain — the abstract reconstruction in particular is the part people forget until they see the raw payload.
If you would rather keep your own client, our residential proxies give a very large crawl stable exits without changing your code.
Search a topic with year_from set to recent years and rank by cited_by_count to surface the works a survey has to cite — with the abstract already readable.
Pull a field's works across years to chart output and citation growth, using venue and type to separate journals from preprints and reviews.
Filter your results on is_open_access and follow oa_url to build a corpus of legally readable full texts for downstream NLP.
One run delivers up to 200 works. Reconstructed abstracts depend on OpenAlex holding an inverted index for the work; some records have none and come back with a null abstract. Runs parallelise up to your plan's rate limit (60–1,200 requests/minute by tier).
OpenAlex is an open dataset under a permissive license; individual full texts at oa_url keep their own licenses. This is not legal advice.
OpenAlex is free and keyless, so use it directly for small queries. This collector adds reconstructed abstracts, a flat schema and rotation for big pulls, plus a $2/month free allowance — around 5,000 papers — with no card.
Yes, and a practical one. Google Scholar has no public API and blocks automated access; OpenAlex indexes the same kind of scholarly record — citations, venues, DOIs, abstracts — through an interface designed to be queried.
Usually. OpenAlex stores abstracts as an inverted index; the collector rebuilds that into readable text in the abstract field. Works without a stored index return a null abstract rather than a fabricated one.
Yes. Set year_from to a year and the search drops anything published earlier, which is the usual way to keep a review focused on current work.