GitHub - TauricResearch/TradingAgents: TradingAgents: Multi-Agents LLM Financial Trading Framework
3 hours ago
- #Multi-Agent Systems
- #AI Trading
- #Open Source
- TradingAgents is an open-source multi-agent LLM financial trading framework designed to simulate real-world trading firms with specialized agents for market analysis and decision-making.
- The framework supports multiple LLM providers including OpenAI, Google, Anthropic, xAI, DeepSeek, Qwen, GLM, OpenRouter, Ollama, and Azure OpenAI, with features like structured-output agents, LangGraph checkpoint resume, and persistent decision logs.
- Key agents include Fundamentals Analyst, Sentiment Analyst, News Analyst, Technical Analyst, Research Manager, Trader, and Risk Management Team, which collaborate to evaluate market conditions and inform trading decisions.
- Installation can be done via cloning the repository, setting up a virtual environment, and installing dependencies, or using Docker, with configuration options for various API keys and enterprise providers.
- The framework includes an interactive CLI for selecting tickers, analysis dates, and LLM providers, and allows programmatic usage through Python code with customizable configurations for model selection and debate rounds.
- State persistence features include a decision log for reflecting on past trades and opt-in checkpoint resume for recovering from interrupted runs, enhancing reliability and learning from historical performance.
- Contributions are encouraged, with credit given in release changelogs, and the project is intended for research purposes, noting that trading performance depends on various factors and is not financial advice.