Local Deep Research
Open-source AI research assistant that runs locally, supports any LLM and search engine, builds a searchable encrypted knowledge base, and generates cited reports automatically.
Smart Download
Visit Project Homepage
No installer available yet — head to the source repository
Local, privacy-first AI research assistant that automates deep research and builds your own knowledge base.
Core Features
- Supports any LLM (Ollama, OpenAI, Anthropic, etc.) and multiple search engines
- 20+ research strategies including LangGraph autonomous agent mode
- Encrypted database; save research sources into your library for future queries
- Fully local, no data leaves your machine
- Docker (CPU/GPU) and pip installation, cross-platform
What It Can't Do
- •SQLCipher encryption may need fallback env var LDR_BOOTSTRAP_ALLOW_UNENCRYPTED=true; 2. Requires 8GB+ RAM (recommended 16GB) for local LLMs; 3. Best search results require SearXNG; others need API keys; 4. On Windows pip install, PDF export needs Pango installed separately.
Use Cases
- Academic researchers: auto-search arXiv, PubMed, Semantic Scholar for literature reviews
- Knowledge workers: build a personal knowledge base from reports and web pages, query with natural language
- Privacy-conscious users: fully offline, no cloud dependency, encrypted storage
Detailed Introduction
Local Deep Research (LDR) is a private, agentic research tool that automates deep research across the web, academic papers, and your own documents. It supports multiple LLMs (via Ollama, OpenAI, etc.), 20+ research strategies, and a new LangGraph agent mode that autonomously decides what to search and when to synthesize. All data is stored in an encrypted SQLCipher database, and you can build a compounding knowledge base by downloading sources. It runs on Docker or pip install, with optional GPU acceleration.
Troubleshooting & FAQ (1)
TroubleshootingWhy does my browser tab crash or run out of memory when the AI chat sends a very long streaming response?
This could be due to an unbounded buffer issue in the chat stream client. Fixed by capping the accumulated streamed content at 256 KB (the same cap used on the server). Once exceeded, the UI shows a truncation notice instead of crashing. Also, the server-side carry buffer for partial citation tokens (e.g., [12) was capped at 64 bytes to prevent unbounded growth. Both fixes are in PR #2953 (commit 3f7a5a73a). Update your deployment to include these changes.
Tags
Getting Started
Download installer
Click the button above to download the installer for your system
Install the software
Double-click the downloaded installer and follow the prompts
Docker: run Ollama, SearXNG, then LDR container (see README for commands)
pip install: pip install local-deep-research, then access localhost:5000
Configure LLM and search engine in settings, choose a strategy, start researching
- Docker: run Ollama, SearXNG, then LDR container (see README for commands)
- pip install: pip install local-deep-research, then access localhost:5000
- Configure LLM and search engine in settings, choose a strategy, start researching
Checksum not available
This project has not published a SHA-256 checksum on its GitHub Release page
SHA256 Checksum
No checksum available
Download directly from GitHub Releases and verify file integrity yourself
All SHA-256 checksums on this platform are extracted from the project's official GitHub Release page, without any modification. You can independently verify them on the GitHub Releases page.
Open Source Transparency
View GitHub SourceUninstall Info
Docker: docker stop local-deep-research && docker rm local-deep-research. pip: pip uninstall local-deep-research.
No Extra Dependencies
Ready to use after download. No additional runtime required.
Having issues? Check the FAQ below
1 FAQ
Similar Projects
goose
Goose is a general-purpose AI agent that runs on your machine. Use it for coding, research, writing, automation, and more. It offers a native desktop app (macOS, Linux, Windows), a full CLI, and an API. Works with 15+ providers (Anthropic, OpenAI, Google, Ollama, etc.) and connects to 70+ extensions via the Model Context Protocol. Built in Rust, part of the Linux Foundation.
Chatbox
Chatbox Community Edition is an open-source desktop client for interacting with multiple large language models. It supports OpenAI (ChatGPT), Azure OpenAI, Claude, Google Gemini Pro, Ollama (local models like Llama 2, Mistral), and ChatGLM-6B. All your chat data is stored locally on your device, ensuring privacy and preventing data loss. The app features a clean, ergonomic UI with dark mode, keyboard shortcuts, streaming replies, and full Markdown/LaTeX rendering with code highlighting. It also includes a prompt library, message quoting, and team collaboration for sharing API resources. Available on Windows, macOS, Linux, Web, iOS, and Android. The community edition is fully functional but may lack some advanced features from the pro version.
daily_stock_analysis
An open-source AI stock analysis system for A/H/US markets that generates daily decision dashboards and pushes them to WeChat Work, Feishu, Telegram, Discord, Slack, or email. Deploy via GitHub Actions for free.