langchain-ai/local-deep-researcher

Fully local web research and report writing assistant

What it solves

Local Deep Researcher automates the process of conducting comprehensive web research on a given topic. It eliminates the need for manual searching and synthesis by iteratively gathering information, identifying knowledge gaps, and refining its search queries to provide a final, well-cited markdown summary.

How it works

The project uses a LangGraph-based workflow that runs entirely on your local machine. It connects to local LLMs via Ollama or LMStudio to perform a multi-step research loop: it generates search queries, retrieves web results, summarizes the findings, and reflects on the summary to determine if further research is needed. This cycle repeats for a user-defined number of iterations to ensure the final output is thorough.

Who it’s for

This tool is for developers and researchers who want to automate information gathering while keeping their data and LLM processing private on their own hardware. It is particularly useful for those already using LangGraph who want to build or experiment with agentic research workflows.

Highlights

  • Fully local execution using Ollama or LMStudio for privacy and control.
  • Iterative research loop that reflects on knowledge gaps to improve results.
  • Built on LangGraph, allowing for clear visualization of the research process.
  • Supports multiple search providers including DuckDuckGo, Tavily, SearXNG, and Perplexity.
  • Produces a final markdown report with citations for all gathered sources.

Related

  • Project
    TheBlewish/Automated-AI-Web-Researcher-Ollama

    An automated research assistant that uses local LLMs via Ollama to systematically search, scrape, and summarize web information based on prioritized research focus areas.

  • Project
    dezoito/ollama-grid-search

    A tool for automating the and evaluation of multiple LLMs, prompts, and inference parameters via Ollama to find the optimal configuration for a specific task.

  • Dispatch
    Building LLM-Powered Web Apps with Client-Side Technology

    Ollama explores the feasibility of building fully local Retrieval-Augmented Generation (RAG) web applications using a combination of browser-based tools and Ollama for local LLM execution.

  • Project
    nashsu/llm_wiki

    A desktop application that automatically turns documents into a structured, interlinked personal wiki using LLMs, featuring a knowledge graph and automated deep research.

  • Project
    taketwo/llm-ollama

    A Python plug‑in for the LLM CLI that discovers and talks to Ollama models (local, remote, or cloud). It adds commands for listing models, prompting, chatting, image attachments, tool use, embeddings, reasoning traces, JSON schemas, and async calls, all configurable via environment variables.