Cito

Hybrid academic search over 236M papers, built for agents

Vercel DayDeveloper ToolsArtificial IntelligenceSearch

▲ 0 votes2 commentsLaunched Jul 16, 2026

Visit Website

Daily #10Weekly #74

Cito is a hybrid search engine over the Semantic Scholar corpus: 236M papers in the keyword index, 146M with SPECTER2 dense vectors, fused with RRF and reranked by a cross-encoder. Free web search with no signup, a plain JSON API, and a native MCP endpoint so agents like Claude Code can run deep literature research without upstream rate limits. Built because every academic API throttled my agents to death.

AI Analysis

📝 Summary

Cito is a hybrid academic search engine over 236M Semantic Scholar papers, combining keyword search with SPECTER2 dense vectors, fused via RRF and reranked by cross-encoder for accurate results. Core features include free web search with no signup, a plain JSON API, and native MCP endpoint optimized for AI agents like Claude to conduct deep literature research without rate limits. It solves key pain points of throttled academic APIs that hinder agent workflows and poor relevance in traditional searches. The value proposition is enabling seamless, high-quality scientific discovery for both human researchers and autonomous AI systems.

📈 Market Timing

In 2025-2026, AI agents and autonomous research tools are experiencing explosive growth, with increasing demand for specialized, rate-limit-free knowledge retrieval to reduce hallucinations and enable deep research. Hybrid search technologies are mature, and developer needs for agent-friendly academic tools align with the rise of MCP and agentic AI frameworks. Economic push for AI productivity tools further supports adoption. This is an Excellent Timing.

✅ Feasibility

Technical difficulty is medium-high due to hybrid search implementation and maintaining relevance at 236M scale, but leverages existing open models (SPECTER2) and Semantic Scholar corpus, reducing data acquisition risks. Vector hosting and inference costs are notable but manageable with cloud providers. No major supply chain issues; compliance is standard for search engines. Strong scalability potential for API usage. Overall rating: High, especially for teams experienced in IR/ML.

🎯 Target Market

Main target segments: AI/ML engineers building research agents (indie devs to enterprise teams), academic researchers, PhD students and STEM faculty. Industries: Artificial Intelligence, Higher Education, R&D labs. Geographic: Global with heavy concentration in US, Europe, and East Asia. Estimated TAM for AI research tools ~$10B+, SAM for academic search ~$1B, SOM for agent-optimized APIs ~$100-200M. Core pains: API throttling, irrelevant results, time sinks in literature review. Strong willingness to pay for reliable, high-throughput API plans.

⚔️ Competition

Competition Level: Medium. Direct competitors: 1. Semantic Scholar (semanticscholar.org), 2. Elicit (elicit.com), 3. Consensus (consensus.app), 4. Scite (scite.ai), 5. Perplexity AI academic search. Advantages: Purpose-built for AI agents with MCP endpoint and no upstream rate limits, superior hybrid fusion/reranking, completely free web access. Disadvantages: Newer entrant with less brand trust, dependency on Semantic Scholar data, potentially higher operational costs without established scale compared to larger players.

Upgrade Pro to unlock full AI analysis