local-deep-research
~95% on SimpleQA (e.g. Qwen3.6-27B on a 3090). Supports all local and cloud LLMs (llama.cpp, Ollama, Google, ...). 10+ search engines - arXiv, PubMed, your private documents. Everything Local & Encrypted.
local-deep-research is an MCP server that enables AI assistants to perform deep, multi-source research with proper citations. It combines agentic search across 10+ engines (arXiv, PubMed, Wikipedia, SearXNG, GitHub) with local or cloud LLMs to generate comprehensive research reports. The server supports both quick summaries and detailed analysis, with all data encrypted locally using SQLCipher. Achieves ~95% SimpleQA accuracy using local models like Qwen3.6-27B on consumer GPUs.
Key Features
Use Cases
- 01Academic literature review across arXiv, PubMed, and Semantic Scholar from within Claude
- 02Privacy-focused research with fully local LLMs and encrypted data storage
- 03Multi-turn research conversations where Claude builds on accumulated context
- 04Private document analysis by indexing company docs or personal knowledge bases
- 05Quick fact-checking with raw search results from Wikipedia, GitHub, or news sources
- 06Comprehensive research reports with structured sections and citation tracking
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
local-deep-research — FAQ
What is local-deep-research?+
An MCP server that gives AI assistants like Claude Desktop and Claude Code the ability to perform deep research across academic databases, web search, and private documents, with all processing done locally or via your chosen LLM provider.
How do I install and connect it to Claude Desktop?+
Install with 'pip install local-deep-research[mcp]', then add the 'ldr-mcp' command to your claude_desktop_config.json with environment variables for your LLM provider (e.g. LDR_LLM_PROVIDER=ollama). Restart Claude Desktop to load the server.
Which AI clients does it work with?+
Works with Claude Desktop and Claude Code via STDIO transport. Can also be used via REST API for integration with other clients.
Do I need API keys?+
Only if using cloud LLM providers (OpenAI, Anthropic, Google) or premium search engines (Tavily, Google via SerpAPI). For fully local operation, you can use Ollama/llama.cpp with SearXNG and no API keys are required.
Is it free and open source?+
Yes, it's MIT licensed and free to use. LLM costs depend on your provider choice—local models via Ollama are free, cloud providers charge per token.
What are the system requirements?+
Requires an AVX-capable CPU (Intel Sandy Bridge / AMD Bulldozer 2011 or newer). For local LLM inference, a GPU like RTX 3090 is recommended but not required; CPU-only mode works with cloud providers.
How do I install local-deep-research?+
Open the source repository on GitHub and follow its README. local-deep-research is a mcp server — MCP Agents Market links you directly to the official repo.
Is local-deep-research free?+
local-deep-research is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.