FinGenius is an open-source AI financial analysis platform designed for the A-share market, offering a multi-agent system that simulates collaborative research and debate among specialized AI analysts. It aims to provide A-share investors with more accurate, context-aware financial insights than generic LLM applications, acting like an experienced financial analyst.

How It Works

FinGenius employs a "Research-Battle" dual-environment architecture. In the Research environment, multiple specialized agents (e.g., sentiment, hot money, risk control) analyze financial data and market trends, leveraging Large Language Models and professional financial tools via the MCP protocol. The Battle environment then facilitates a structured, multi-round debate among these agents, incorporating game theory principles to refine decision-making, before presenting a consolidated HTML report. This approach aims to mitigate LLM "hallucinations" and provide persistent, context-aware analysis.

Quick Start & Requirements

• Installation: Recommended via uv (a fast Python package manager).
  1. Install uv: curl -LsSf https://astral.sh/uv/install.sh | sh
  2. Clone repo: git clone https://github.com/HuaYaoAI/FinGenius.git && cd FinGenius
  3. Create venv: uv venv --python 3.12 && source .venv/bin/activate
  4. Install dependencies: uv pip install -r requirements.txt
• Configuration: Requires setting up an LLM API key in config/config.toml (e.g., OpenAI).
• Execution: Run analysis with python main.py <stock_code>.
• Prerequisites: Python 3.12, uv (optional but recommended), LLM API access.

Highlighted Details

• Features 16 specialized AI agents (6 currently open-sourced: sentiment, hot money, risk control, technical analysis, chip analysis, large order anomalies).
• Utilizes a "Research-Battle" environment for sequential analysis and structured debate among agents.
• Incorporates a proprietary "annual ring memory rule algorithm" for persistent, habit-aware user context.
• Supports text-to-speech output and customizable debate rounds.

Maintenance & Community

The project is actively seeking contributors and invites developers to join the FinGenius open-source community. Contact information or community links are not explicitly provided in the README.

Licensing & Compatibility

FinGenius is licensed under GPL v3. The project explicitly states that closed-source commercial use of the code and framework is prohibited, and they reserve the right to pursue legal action.

Limitations & Caveats

The project is intended for educational and research purposes only and does not constitute investment advice. The mobile application is currently available on Huawei, Xiaomi, and Vivo app stores, with Apple and other platforms pending review.