Semantic paper search
Runs semantic retrieval using SPECTER2 embeddings trained for scientific papers, so queries can match papers by meaning even when the wording differs.
Cito
Cito is a semantic search service for academic papers. Find research by meaning, with keyword and hybrid search, structured metadata, saved papers, and API keys.
Overview
Cito is a semantic search service for academic papers. It is designed to help people find research by meaning rather than by exact keywords, and it covers a corpus of roughly 148 million papers from Semantic Scholar and OpenAlex data.
The product exposes a programmatic API and search interface that combine semantic retrieval with keyword matching. Queries can be run against Cito's own self-hosted indexes, which makes results deterministic for the same query and corpus snapshot and avoids relying on upstream API rate limits at search time.
Core features
Hybrid ranking
Combines classic BM25 full-text matching over titles and abstracts with semantic retrieval, then merges the rankings with reciprocal-rank fusion.
Selectable search modes
Can run either retriever on its own with mode=semantic or mode=keyword, which lets users narrow the search strategy when needed.
Self-hosted corpus and query path
Searches its own self-hosted indexes rather than calling a third-party API at query time, which keeps results reproducible and avoids inherited upstream rate limits.
Metadata and paper links
Returns structured paper metadata, including title, authors, venue, publication date, DOI, citation count, abstract when available, and an open-access PDF link when one exists.
Account access
Supports user accounts for saved papers, API keys, and usage viewing through GitHub sign-in or an existing API key.
Common use cases
Literature discovery by concept
Researchers can start from a landmark paper or a concept and surface related work even when the relevant papers use different terminology.
Programmatic paper search
Developers can query the API from their own tools or workflows when they need programmatic access to academic search results and metadata.
Embedding academic search in apps
Teams building research-facing products can reuse the searchable corpus and response metadata in commercial contexts, while keeping the required attribution string in republished results.
Exact-term and acronym lookups
Users who need exact terms, acronyms, or author jargon can switch to keyword search, or use the hybrid mode to balance precise matching with semantic recall.
Pros and Cons
Pros
- Searches by meaning as well as by exact terms, which can help surface relevant work that does not share keywords with the query.
- Merges semantic and keyword retrieval, preserving both conceptual matches and exact jargon or acronyms.
- Uses a self-hosted search path, so it does not inherit the public Semantic Scholar API's 1 request/second limit at query time.
- Returns practical paper metadata and open-access PDF links when available.
- Licensing page states the underlying data can be used in commercial products, research, and publications with attribution.
Cons
- The public site provides limited detail on pricing; the pricing page currently returns a 404 in the collected evidence.
- Full text is not hosted by Cito, so access to PDFs depends on the originating publisher or repository.
FAQ
What kind of search does Cito provide?
Cito is a search service for academic literature. It supports semantic search, keyword search, or a fused search mode that combines both retrievers.
What information does a result include?
The source documentation says Cito returns title, authors, venue, publication date, DOI, citation count, an abstract when available, and a direct open-access PDF link when one exists.
Can Cito be used commercially?
Yes. The licensing page says Cito can be used in commercial products, research, and publications, with the requirement to keep the attribution string that every API response carries: “Data from Semantic Scholar” when republishing results.
Does Cito support user accounts or API keys?
Cito has an account area for signing in with GitHub, saving papers, creating API keys, and viewing usage. It also says an API key can be used to view your account.
Does Cito host the full papers?
Cito does not proxy or re-host full text. PDF links point to the original publisher or repository, and access to the PDF is governed by that source's terms.
Quick Facts
- Category
- Developer Tool
- Primary use
- Semantic search over academic papers
- Search corpus
- About 148 million papers
- Search modes
- Semantic, keyword, and hybrid
- Auth / account
- GitHub sign-in and API keys
- Source domain
- cito.fim.ai
Cito Alternatives
Scite
Scite is an AI research platform for finding papers, checking citation context, and getting answers grounded in scholarly literature and patents.
Model Council
Model Council on Perplexity runs one query across multiple AI models and combines the results into one answer. Available to Perplexity Max on the web.
GoldenRetriever
GoldenRetriever is a macOS public beta for AI search across videos, audio, and documents in your local media library, with sourced answers and timestamps using your own Gemini API key.
Hype
Hype is a web tool for finding trending YouTube topics by category, time range, and scoring mode. Spot emerging ideas and review source videos.
Falconer
Falconer is an AI-powered company brain that keeps internal docs accurate, searchable, and up to date by syncing with GitHub, Slack, and Linear.
ClayHog
ClayHog is an AI search visibility platform for tracking brand presence in ChatGPT, Gemini, Perplexity, Claude, and Google AI Overviews from one dashboard.