Semantic Scholar
Free AI-powered search engine for scientific literature, backed by a nonprofit.
About
Semantic Scholar is a research tool for scientific literature developed at the Allen Institute for AI (AI2) and publicly released in November 2015. It uses natural language processing to support discovery: automatically generated paper summaries, influential-citation identification, and semantic search over a corpus that passed 200 million publications across all fields of science as of September 2022. The platform also offers Semantic Reader, an augmented PDF reader with in-line citation cards and TL;DR summaries, plus personalized research feeds and a public REST API with Academic Graph, Recommendations, and Datasets endpoints.
It is one of the few research tools with a genuinely free model. AI2 states there are no plans to charge for it, and it is funded by philanthropist Paul G. Allen's estate, so there are no paid tiers or premium features. Its influence runs deeper than the website itself: downstream literature tools like Elicit and Consensus build on its corpus and citation data, which makes it shared infrastructure for AI research discovery. That dependency is worth knowing, since a paper missing from its index is invisible to every tool built on top of it.
For researchers the tradeoffs are coverage versus polish. Google Scholar still reaches further into older papers and books, but Semantic Scholar tends to return fewer, higher-relevance results with AI summaries attached, especially in computer science and biomedicine. The vendor presents citation-influence classification as a core advantage over keyword search, and independent technical writeups broadly agree that the citation-graph approach is the real differentiator.
What it can do
- AI semantic search over 200M+ papers across all scientific fields
- Automatically generated TL;DR paper summaries
- Influential citation identification and citation context
- Semantic Reader augmented PDF reading with in-line citation cards
- Personalized research feeds and email alerts
- Free public REST API: Academic Graph, Recommendations, Datasets
Setup
- Go to semanticscholar.org and search immediately, no account needed.
- Create a free account to save papers to a library and set up alerts.
- Open papers in Semantic Reader for in-line citation cards and TL;DR summaries.
- Build a research feed from a folder or topic to get personalized recommendations.
- For programmatic use, request a free API key for the Academic Graph or Datasets APIs.
Prerequisites
None for search; a free account for libraries, alerts, and feeds.
Use cases
Works with
Trust score — 9/10
Scored 9/10 (verified tier, metrics-v1). Evidence: launched Nov 2, 2015 by Allen Institute for AI; 200M+ publications indexed as of Sept 2022 (Wikipedia); nonprofit funded by Paul G. Allen's estate, no paid tiers planned (costbench.com). Docs corroboration: free public API documented at api.semanticscholar.org; Semantic Reader and research feeds documented by bestaitools.com and moge.ai. Strengths: nonprofit permanence, genuine infrastructure role (Elicit and Consensus build on its corpus per technologynetworks.com), strongest documentation and API story in the category. Docked for: (1) coverage still trails Google Scholar for older papers and books; (2) summary and relevance quality depend on AI accuracy, which varies by field. License: free nonprofit service; API free for academic and research use.
How the trust score is built
Full methodology →- Adoption · stars/downloads, log-scaled vs peers
- 10.0/10
- Maintenance · recency, releases, issue load
- 9.0/10
- Openness · license + docs
- 10.0/10
- Manual review · sanity read + red-flag check
- 9.0/10
- · Launched Nov 2, 2015 by Allen Institute for AI; 200M+ publications indexed as of Sept 2022 (Wikipedia)
- · Nonprofit funded by Paul G. Allen's estate; no paid tiers or plans to charge (costbench.com)
- · Free public API (Academic Graph, Recommendations, Datasets) at api.semanticscholar.org; Elicit and Consensus build on its corpus (technologynetworks.com; github.com api-evangelist)
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.