ArXiv Research MCP Server
Full-text search across 2M+ ArXiv papers with category filtering, retrieve paper details including citations and code repos, and get trending papers by category.
Find GitHub code implementations for papers and look up state-of-the-art benchmarks by task or dataset.
Citation analysis with citing papers and references, and semantic search better than keyword search.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
ArXiv Research MCP Server π¬
The MCP server every data scientist needs. Search 2M+ ArXiv papers, find code implementations, track citations, get SOTA benchmarks β all from your AI assistant.
Why This Server?
Data scientists spend ~30% of their time searching for papers, checking if code exists, comparing benchmarks. This MCP server brings all of that into your LLM workflow with zero API keys required.
Related MCP server: arxiv-query-mcp
Tools Available
Tool | What it does |
| Full-text search across 2M+ ArXiv papers with category filtering |
| Complete paper details with citations + code repos enrichment |
| Citation analysis via Semantic Scholar (citing papers, references) |
| Find GitHub repos implementing a paper (Papers with Code) |
| Latest papers in any ArXiv category |
| SOTA benchmark lookup by task or dataset |
| Semantic search via Semantic Scholar (better than keyword search) |
| List all supported ArXiv categories |
Quick Start
Option 1: Connect directly (Streamable HTTP)
Add to your Claude Desktop / MCP client config:
{
"mcpServers": {
"arxiv-research": {
"type": "url",
"url": "http://localhost:8000/mcp"
}
}
}Option 2: Run locally
pip install -r requirements.txt
python arxiv_research_mcp.pyOption 3: Docker
docker build -t arxiv-research-mcp .
docker run -p 8000:8000 arxiv-research-mcpOption 4: Using uv (recommended)
uv run --with "mcp[cli]" --with httpx --with pydantic arxiv_research_mcp.pyExample Queries
Once connected, ask your AI assistant:
"Search ArXiv for papers on LoRA fine-tuning from 2024"
"Get full details for paper 2106.09685 including citations and code"
"Find code implementations for the Mamba architecture paper"
"What are the trending papers in cs.CL this week?"
"Show SOTA benchmarks for object detection"
"Semantic search for papers on retrieval augmented generation"
APIs Used (All Free, No Keys)
API | Rate Limit | Data |
ArXiv API | 1 req/3s | 2M+ papers, full metadata |
Semantic Scholar | ~100 req/5min | Citations, references, TL;DR |
Papers with Code | Fair use | Code repos, benchmarks, tasks |
Deployment
Railway / Render / Fly.io
The server runs on port 8000 with streamable HTTP transport. Deploy the Dockerfile to any container hosting platform. No environment variables or API keys needed.
Permanent self-hosting
docker compose up -dLicense
MIT β use it, fork it, make it better.
Author
Dr. Swapnil Morande β Principal AI Architect
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelogβ¦
This server cannot be deployed
Maintenance
Related MCP Connectors
Research paper search with real citations and reference formatting for AI assistants
Research intelligence for AI coding agents. 2M+ CS papers with evidence and tradeoffs.
Paper search (free), literature review packs, hallucinated-citation checks, arXiv digests, citations
Search 340M+ academic papers β citation graphs, semantic similarity, and AI literature reviews.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables natural language search and analysis of arXiv academic papers with AI-powered relevance ranking, full-text extraction, and support for multiple integrations like Claude and LangChain.MIT
- AlicenseAqualityBmaintenanceEnables AI assistants to search arXiv papers, retrieve metadata, browse categories, and read paper text.5MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to discover and manage arXiv papers by searching for papers, extracting metadata, and browsing saved collections, all within the conversation.1MIT
- AlicenseNot gradedqualityDmaintenanceEnables AI assistants to search, download, and read arXiv papers, with automatic detection of open-source code repositories and support for both LaTeX and PDF content.MIT