OpenLens AI is a fully autonomous multimodal research agent designed to automate complex data-driven projects, particularly in medical, ML, and statistical research. It empowers users to provide a dataset and a research idea, after which the agent independently handles literature review, experiment design, data analysis, and report generation without manual intervention. Its key benefit is streamlining the entire research lifecycle into a single, automated process.

How It Works

The system employs a multi-module architecture orchestrated by LangGraph. Core agents include a Literature Reviewer, Data Analyzer, Supervisor, Coder, and LaTeX Writer. These agents communicate via a shared state, leveraging tools such as Tavily for web search, OpenHands for code execution within a Docker sandbox, and vector search for context management. This modular, agent-based approach allows for flexible and robust automation of diverse research tasks.

Quick Start & Requirements

• Primary Install/Run: Clone the repository (git clone --recurse-submodules), set up a Conda environment (Python 3.12 recommended), install dependencies (pip install -e .), and configure API keys in config.toml. Run via CLI (python cli.py) or Streamlit UI (streamlit run start_app.py).
• Prerequisites: Python 3.9+, Docker, API keys for LLM services (e.g., DeepSeek, OpenAI), Tavily search API, and optionally reranking API. For Chinese language support, downloading specific fonts may be required.
• Links: 📄 Paper on arXiv, 🌐 Project Page, 🚀 Cloud Application.

Highlighted Details

• Automated literature review supporting arXiv, medRxiv, Google Scholar, and Tavily search.
• Comprehensive data analysis with report generation.
• Experiment design suggestion and validation capabilities.
• Code generation and execution via the OpenHands sandbox.
• Automated LaTeX paper generation and management.
• Interactive Streamlit-based UI for monitoring.
• Support for general domains beyond medical research.
• Full Chinese language support for figures and paper writing.
• Vision-Language Model integration for visualization and feedback.

Maintenance & Community

No specific details regarding notable contributors, sponsorships, or community channels (like Discord/Slack) were found in the provided README.

Licensing & Compatibility

This project is licensed under the MIT License, which is permissive and generally compatible with commercial use and closed-source linking.

Limitations & Caveats

The setup requires obtaining and configuring multiple third-party API keys, which can be a barrier to entry. The installation process involves Docker and Conda environments, potentially adding complexity. Some advanced literature search tools are marked as optional or not yet implemented. Full Chinese language support may necessitate manual font downloads.