PubMed Research Scraper

Search 37M+ PubMed biomedical articles via the NCBI E-utilities API and get clean, typed JSON output. One chargeable row per article — pay only for what you receive.

Why this actor?

• Real PubMed data — queries live against the NCBI E-utilities API, not a stale mirror.
• Structured output — flat typed JSON with title, authors, journal, pubdate, DOI, and article URL.
• Full query syntax — supports boolean operators, field tags ([ti], [au], [mh]), date range filters, and pagination.
• Fair per-article pricing — $0.002 per delivered article. Empty results and errors are free.
• Agent-ready — one row per article, no pagination nesting. Drop the JSON into any LLM or MCP pipeline.

Input

{
  "query": "cancer immunotherapy",
  "maxResults": 20,
  "start": 0,
  "dateFrom": "2024/01/01",
  "dateTo": "2026/12/31"
}

Field	Type	Required	Default	Description
query	string	yes	—	PubMed search query (supports boolean, field tags, phrases)
maxResults	integer	no	20	Maximum articles (1–200)
start	integer	no	0	Pagination offset
dateFrom	string	no	—	Filter from date (YYYY/MM/DD)
dateTo	string	no	—	Filter to date (YYYY/MM/DD)

Output

One item per article:

{
  "input": "cancer immunotherapy",
  "pmid": "39999999",
  "title": "Advances in Cancer Immunotherapy",
  "authors": ["Smith J", "Doe K"],
  "journal": "Nat Med",
  "pubdate": "2025 Jan 15",
  "doi": "https://doi.org/10.1038/s41591-025-99999-9",
  "abstract": null,
  "articleUrl": "https://pubmed.ncbi.nlm.nih.gov/39999999/",
  "error": null,
  "scraped_at": "2026-08-03T12:00:00.000Z"
}

Pricing

Pay per event, and the platform usage is on us — the price you see is the price you pay, with no compute bill on top.

Event	Price	What one charge buys
PubMed article result	$0.002	Charged for each PubMed article returned by the search. Empty results are never charged. Errors and parse failures are free.
Actor Start	$0.002	One run, whatever it returns.

Higher Apify subscription tiers pay less on every event (Silver −20%, Gold −35%).

Use cases

• Systematic reviews — search PubMed with structured queries and export clean JSON for meta-analysis.
• Competitive intelligence — track publications by author, institution, or MeSH term across date ranges.
• LLM RAG pipelines — feed PubMed abstracts and metadata into retrieval-augmented generation workflows.
• Bibliometric dashboards — collect publication counts and metadata for research analytics.

FAQ

What query syntax does PubMed support? The full NCBI PubMed query syntax: boolean operators (AND, OR, NOT), field tags (cancer[ti] for title search, smith j[au] for author), MeSH terms (diabetes mellitus[mh]), date ranges (2024[dp]), and phrase search with double-quotes.

How many articles can I get per run? Up to 200, in increments of 1. Use start for pagination across multiple runs.

Is there a rate limit? NCBI allows 3 requests/second without an API key. We pace at ~2.86 req/s (350ms between calls) to stay within limits.

Are abstracts included? Not by default (to keep the response fast and lightweight). Abstracts require an extra NCBI efetch call per article and will be added in a future version.

---

This Actor is an independent tool built on the public NCBI E-utilities API. It is not affiliated with, endorsed by, or sponsored by the National Center for Biotechnology Information (NCBI), the U.S. National Library of Medicine (NLM), or the National Institutes of Health (NIH).