In simple English
This tool acts as an automated research assistant on your computer. You give it any topic, and it writes search queries to explore the web. It reads the results, finds what is missing, and searches again to fill the gaps. It repeats this cycle several times until it has enough material. Finally, it gives you a complete summary with all the sources attached.
What it is
A Python application built on the LangGraph framework that coordinates an artificial intelligence model and a web search tool. It runs locally alongside software like Ollama (a tool that lets you run large language models on your own machine) or LM Studio.
What it is good for
Solo builders who want to research competitors, market trends, or technical topics without sending sensitive data to cloud services or paying per query.
- Runs completely offline with your own language models, keeping your data private.
- Uses DuckDuckGo by default so you do not need paid search keys to get started.
- Generates structured markdown reports with full source citations automatically.
- Requires a powerful computer with plenty of RAM or a graphics card to run large local models smoothly.
- Setup involves installing multiple moving parts like Python, Ollama, and a development server.
- Smaller language models may struggle with the structured format the assistant requires.
Manual web browsing, copy-pasting search results into chat windows, and writing research notes by hand.
Ollama (a tool that runs language models locally on your computer) to provide the local intelligence.
- HKUDS/LightRAG40k stars
Another local research and retrieval system, focusing on graph-based indexing instead of iterative search loops.
- assafelovic/gpt-researcher30k stars
The broader cloud-based research agent framework that inspired many local web research tools.
- ggml-org/llama.cpp132k stars
A C and C++ library and command-line tool for running language models locally.
- EleutherAI/lm-evaluation-harness14k stars
Runs standard question sets against a language model and reports the scores.
- mudler/LocalAI50k stars
Serves text, vision, speech, image and video models from one self-hosted service that imitates the paid interfaces.
git clone https://github.com/langchain-ai/local-deep-researcher.git cd local-deep-researcher cp .env.example .env python -m venv .venv source .venv/bin/activate pip install -e . pip install -U "langgraph-cli[inmem]" langgraph dev