# TauricResearch/TradingAgents

TradingAgents: A Multi\-Agent LLM Financial Trading Framework

- HackerByte page: <https://hackerbyte.net/en/projects/tradingagents-a-multi-agent-llm-financial-trading-framework-909213664>
- GitHub: <https://github.com/TauricResearch/TradingAgents>
- Programming language: Python

TradingAgents: Multi\-Agents LLM Financial Trading Framework

## AI-generated summary

TradingAgents is an open\-source framework designed to simulate real\-world trading firms using multiple Large Language Model \(LLM\) powered agents. It breaks down complex financial trading tasks into specialized roles, such as analysts, researchers, traders, and risk managers, who collaborate to evaluate market conditions and make informed trading decisions. The framework supports a wide range of LLM providers and offers both a command\-line interface \(CLI\) and Python package for integration.

## Key points

- Employs specialized LLM agents \(Fundamental Analyst, Sentiment Analyst, Technical Analyst, Trader, Risk Manager\) for collaborative financial trading.
- Supports numerous LLM providers including OpenAI, Google, Anthropic, Mistral, Groq, and local Ollama models.
- Offers a flexible CLI for interactive use or script automation and a Python package for programmatic integration.
- Features persistence through a memory log for tracking past decisions and outcomes, and checkpoint resume for robust execution.
- Provides backtesting capabilities to evaluate agent decisions over a grid of tickers and dates.