AI/ML Financial Trading Framework
76.3k 2026-05-17
TauricResearch/TradingAgents
A multi-agent LLM framework designed to simulate real-world trading firms, enabling collaborative evaluation and informed financial trading decisions.
Core Features
Multi-agent architecture with specialized LLM-powered roles (analysts, traders, risk management).
Dynamic agent discussions to pinpoint optimal trading strategies.
Support for a wide range of LLM providers (e.g., GPT, Gemini, Claude, DeepSeek, Qwen).
Structured output agents for research, trading, and portfolio management.
Backtesting capabilities and persistent decision logging.
Detailed Introduction
TradingAgents is an advanced open-source framework that leverages a multi-agent system powered by Large Language Models (LLMs) to mimic the operational dynamics of professional trading firms. It deploys specialized agents, including fundamental, sentiment, and technical analysts, as well as traders and risk managers, to collaboratively assess market conditions. These agents engage in dynamic discussions to formulate and execute optimal trading strategies, providing a robust platform for research and development in AI-driven financial trading.