Repolode
Submit
TauricResearch

TauricResearch/TradingAgents

Open-source multi-agent LLM framework that simulates a trading firm, with analyst, researcher, trader and risk agents debating market decisions.

View on GitHubWebsite
Stars
111,998
Forks
21,644
Open issues
99
Last commit
4 days ago

What is it?

TradingAgents is a Python framework built on LangGraph that models a trading firm as a set of LLM-powered agents. Fundamentals, sentiment, news and technical analysts each gather data with their own tools, then bullish and bearish researchers debate their reports. A trader proposes a transaction, and risk management and a Portfolio Manager approve or reject it, with approved orders sent to a simulated exchange.

The project supports many model providers, including OpenAI, Google, Anthropic, DeepSeek, Qwen, Ollama and any OpenAI-compatible endpoint, and lets the quick and deep reasoning tiers run on different providers. It offers an interactive CLI, a Python API, Docker setup, checkpoint resume, a persistent decision memory log, and a backtesting module that scores decisions over a grid of tickers and dates. Its data sources cover Yahoo Finance, SEC EDGAR filings and FRED macro data, and it works across several markets.

The framework is intended for research into multi-agent analysis rather than as a trading strategy. Its outputs vary between runs because language model sampling and live data are non-deterministic.

Key features

  • 01Analyst, researcher, trader and risk agents collaborate and debate before a decision.
  • 02Supports OpenAI, Google, Anthropic, DeepSeek, Qwen, Ollama and any OpenAI-compatible endpoint.
  • 03Interactive CLI, Python API and Docker setup.
  • 04Checkpoint resume and a persistent decision memory log.
  • 05Backtesting across a grid of tickers and dates.
  • 06Covers Yahoo Finance, SEC EDGAR and FRED data across multiple markets.

Use cases

  • Researching how multi-agent LLM systems reason about financial decisions.
  • Prototyping agent workflows on LangGraph with real market data.
  • Backtesting LLM-driven analysis across historical dates and tickers.
  • Comparing model providers on the same trading pipeline.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/TauricResearch/TradingAgents.git
  2. 2

    Enter the project folder

    cd TradingAgents
  3. 3

    Create a Python 3.13 environment with conda

    conda create -n tradingagents python=3.13
  4. 4

    Install the package and its dependencies

    pip install .

Things to know

  • Designed for research purposes only and not intended as financial or trading advice.
  • Requires an API key from an LLM provider, or a local model setup such as Ollama.
  • Trading performance may vary with the chosen models, temperature, data quality and other non-deterministic factors.

Topics

#agent#llm#multi-agent#finance#trading#langgraph#backtesting#multiagent

FAQ

What is TradingAgents?
Open-source multi-agent LLM framework that simulates a trading firm, with analyst, researcher, trader and risk agents debating market decisions. TradingAgents is a Python framework built on LangGraph that models a trading firm as a set of LLM-powered agents. Fundamentals, sentiment, news and technical analysts each gather data with their own tools, then bullish and bearish researchers debate their reports. A trader proposes a transaction, and risk management and a Portfolio Manager approve or reject it, with approved orders sent to a simulated exchange.
Is TradingAgents free to use? What license does it have?
TradingAgents is open source under the Apache-2.0 license. Check the license terms before using it in commercial products.
Is TradingAgents still maintained?
Actively maintained. The last commit to TradingAgents was on Oct 16, 2026.
What can I use TradingAgents for?
Researching how multi-agent LLM systems reason about financial decisions. Prototyping agent workflows on LangGraph with real market data. Backtesting LLM-driven analysis across historical dates and tickers. Comparing model providers on the same trading pipeline.
How do I install TradingAgents?
Clone the repository: git clone https://github.com/TauricResearch/TradingAgents.git Enter the project folder: cd TradingAgents Create a Python 3.13 environment with conda: conda create -n tradingagents python=3.13 Install the package and its dependencies: pip install .
Where can I get TradingAgents?
The source code and installation instructions are on GitHub: https://github.com/TauricResearch/TradingAgents

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects