Semantic Scholar Scraper

Bulk-scrape academic papers from Semantic Scholar, an AI-powered corpus of 200M+ papers across every field. Get titles, abstracts, authors, venues, years, citation and influential-citation counts, fields of study, cross-IDs (DOI / PubMed / arXiv) and open-access PDF links. Filter by year, field of study and citation count. Thousands of papers per run. No account, no API key.

Features

• 🔎 Corpus-wide search — 200M+ papers, all disciplines
• 📈 Citation metrics — total citations + influential citations + reference counts
• 🎓 Filters — year range, fields of study, minimum citations, open-access-only
• 🔗 Cross-IDs — DOI, PubMed and arXiv IDs for each paper, plus free PDF links
• 👥 Authors with IDs
• 📊 Bulk & token-paginated — scales to tens of thousands of papers per query

What you get

One record per paper:

{
  "paperId": "649def34f8be52c8b66281af98ae884c09aef38b",
  "title": "…",
  "year": 2023,
  "authors": [{ "name": "…", "authorId": "1741101" }],
  "venue": "NeurIPS",
  "citationCount": 1240,
  "influentialCitationCount": 210,
  "referenceCount": 42,
  "fieldsOfStudy": ["Computer Science"],
  "doi": "10.…",
  "pmid": null,
  "arxivId": "1706.03762",
  "openAccessPdfUrl": "https://…",
  "url": "https://www.semanticscholar.org/paper/…"
}

(abstract included when available.)

Output schema

Input

{
  "query": "graph neural networks",
  "yearFrom": "2020",
  "fieldsOfStudy": ["Computer Science"],
  "minCitationCount": 10,
  "maxResults": 5000
}

Use cases

• Literature reviews — every paper on a topic, ranked by citations
• Citation & impact analysis — find the influential works and authors in a field
• AI/ML datasets — build corpora of titles + abstracts + metadata
• Cross-referencing — map papers across DOI, PubMed and arXiv

Pricing

Pay-per-event: a small amount per paper scraped. See the Pricing tab for current rates.

FAQ

Do I need an API key? No. This Actor uses the public Semantic Scholar Academic Graph API with no account required. You can optionally supply an S2 apiKey for a higher rate limit on very large runs.

How many papers can I scrape per run? Set maxResults as high as you need — the API is token-paginated and scales to tens of thousands of papers per query.

Is scraping Semantic Scholar legal? Yes. This Actor uses the official Academic Graph API and returns its openly available metadata as-is, with no login or scraping tricks.

What format is the output? Structured JSON — one record per paper — exportable as JSON, CSV or Excel. Each record includes title, abstract (when available), authors, venue, year, citation and influential-citation counts, fields of study, cross-IDs (DOI/PubMed/arXiv) and open-access PDF links.

Can I filter by year, field or citation count? Yes. Use yearFrom/yearTo for a publication-year range, fieldsOfStudy to restrict disciplines, minCitationCount to keep only well-cited papers, and openAccessOnly for papers with a free PDF.

Can I integrate this with other tools? The Semantic Scholar Scraper can be connected with almost any cloud service or web app thanks to integrations on the Apify platform. It works with Make, Zapier, Slack, Airbyte, GitHub, Google Drive and many more, plus the Apify API, JavaScript/Python clients and MCP. Or use webhooks to trigger an action whenever a run finishes.

Building a citation or literature dataset? These other hipersoft scrapers pair well with this one:

• OpenAlex Scraper — 250M+ scholarly works with citations, authors and abstracts
• Crossref Scraper — DOIs, citation counts and metadata from 150M+ works
• arXiv Papers Scraper — preprints with full abstracts and PDF links
• PubMed Scraper — biomedical papers, abstracts and MeSH terms

Notes

Uses the public Semantic Scholar Academic Graph API. Returns its openly-available metadata as-is. This is an independent tool and is not affiliated with or endorsed by Semantic Scholar or the Allen Institute for AI.

You pay only for what the run returns. Set a maximum cost per run and it stops cleanly at that limit.