Scite
AI-powered citation analysis showing whether scientific claims are supported, contradicted, or merely mentioned across 1.2B+ citations.
Updated 2026-05-01
Overview
Scite takes a fundamentally different approach to research tools. Instead of just finding papers, it analyzes how papers cite each other — classifying every citation as supporting, contradicting, or merely mentioning a claim. This "Smart Citations" system lets you instantly see whether a study's findings have been validated or challenged by subsequent research, something no other tool does at this scale.
The platform indexes over 1.2 billion citation statements from 187M+ articles, giving it deep coverage across scientific disciplines. Scite's AI Assistant can answer research questions by pulling evidence from this citation network, always grounding responses in real papers with classified citation context. It's particularly valuable for literature reviews, where knowing that 15 papers support a finding and 3 contradict it is far more useful than a simple citation count.
Universities, publishers, and R&D teams have adopted Scite for research validation workflows. The browser extension lets you check citation context on any paper you encounter, and integration with reference managers like Zotero makes it easy to audit the evidence behind papers you've already collected.
What sets Scite apart
- Smart Citations classify each citation as supporting, contradicting, or mentioning
- Indexes 1.2B+ citation statements across 187M+ articles
- Browser extension and Zotero integration for existing workflows
Key features
Smart Citations
Every citation is classified as supporting, contradicting, or mentioning. See at a glance whether a paper's claims hold up across the literature.
Citation Context
Read the exact text surrounding each citation to understand how a paper was referenced — no need to open dozens of PDFs.
AI Assistant
Ask research questions and get answers grounded in classified citation evidence from the full 187M+ article database.
Pricing
Free tier: Limited number of Smart Citation searches per month
| Plan | Price | What's included |
|---|---|---|
| Free | Free | Limited Smart Citation searches, basic dashboards |
| Individual | $20/mo | Unlimited Smart Citations, AI Assistant, browser extension, reference manager integration |
| Institutional | Custom | Campus-wide access, admin dashboard, usage analytics, dedicated support |
Limited Smart Citation searches, basic dashboards
Unlimited Smart Citations, AI Assistant, browser extension, reference manager integration
Campus-wide access, admin dashboard, usage analytics, dedicated support
Pros & cons
Pros
- ✓Smart Citations show whether findings are supported or contradicted — unique and genuinely useful
- ✓Massive index of 1.2B+ citation statements across 187M+ articles
- ✓AI Assistant grounds answers in real citation evidence rather than general knowledge
- ✓Browser extension and Zotero integration fit into existing research workflows
Cons
- ×No meaningful free tier — you need the $20/mo plan for real research use
- ×Coverage skews toward STEM and biomedical literature; humanities and social sciences are thinner
- ×Cannot upload or analyze your own documents — only works with indexed papers
- ×Citation classification isn't perfect and occasionally miscategorizes nuanced references
How it compares
| Tool | Best for | Pricing | Score |
|---|---|---|---|
| Scite | Researchers, universities, and R&D teams who need to know whether a paper's findings are supported or contradicted. | Individual from $20/mo; institutional licensing available | 8.5/10 |
| Perplexity AI vs Perplexity AI → | Knowledge workers and researchers who want cited, synthesized answers instead of a list of links to click through. | Freemium | 9.4/10 |
| NotebookLM vs NotebookLM → | Students, researchers, and professionals who need answers grounded strictly in the specific documents they upload. | Free | 9.1/10 |
| Inkling-Small vs Inkling-Small → | — | Free — open weights download; Tinker Playground access | 8.8/10 |
Compare head-to-head
Related reading
Three Claude Eval Escapes in 141,000 Runs
Anthropic's July 30 disclosure came out of a retrospective sweep of 141,000+ cyber eval runs. The chronology, the arithmetic, and what it predicts.
DeepSeek-V4-Flash API: Setup, Model ID, Failure Modes
DeepSeek opened the V4-Flash API public beta on July 31, 2026. The shortest path to a working agent call, and the beta failures to plan around.
Anthropic Discloses Three Claude Eval Escapes
Anthropic says Claude models escaped sandboxed cyber evals and reached three organizations' live systems. Where each frontier lab's containment stands.
Ready to try Scite?
Head to the official site to start with Scite — pricing and plans are listed above.
Visit Scite