Research Papers API for Agents
Pricing
from $0.01 / 1,000 results
Go to Apify Store
Research Papers API for Agents
Search OpenAlex, arXiv and Crossref in one unified, token-efficient JSON schema built for AI research agents.
Research Papers API for Agents
Pricing
from $0.01 / 1,000 results
Search OpenAlex, arXiv and Crossref in one unified, token-efficient JSON schema built for AI research agents.
The main search keyword(s) or phrase, e.g. "large language models", "CRISPR gene editing". Required. Matched as a phrase/topic search against each requested source (title/abstract/full-text depending on the source).
Which open scholarly API(s) to query. "openalex" (richest: citations, concepts, open-access links, DOI) is the default. Add "arxiv" for preprints (no citation data) and/or "crossref" (strong DOI/venue coverage, abstracts rare). When multiple sources are requested, results are round-robin mixed across them before max_items is applied, not just filled from the first-listed source. Papers found in more than one source are deduplicated by DOI within a run.
[ "openalex"]Only return papers published in or after this year, e.g. 2020. Omit to skip this filter entirely (no lower bound).
Only return papers with at least this many citations. Note: arXiv exposes no citation data at all, so with this filter set, arXiv-sourced results will always be excluded (their citation count is always unknown, not zero). Omit to skip this filter entirely.
If true, only return papers with a usable open-access PDF link (oa_pdf_url non-null). Crossref rarely has a reliable open-access signal (see README), so this filter mostly narrows OpenAlex/arXiv results.
Which output fields to include per paper, for token efficiency (e.g. omit 'concepts' and 'abstract' if you only need titles/authors/citations). Leave empty to return every field. 'source' and 'title' are always included regardless of this setting.
[]Whether to fetch/return the abstract at all. Abstracts are the most token-expensive field in the output — set this to false if you only need titles/metadata. When false, 'abstract' is always null regardless of the 'fields' setting.
Maximum number of UNIQUE papers to return in total, across all requested sources combined (after cross-source DOI deduplication). Must be between 1 and 200. Each paper returned bills one 'paper-result' event, so lowering this caps cost.