TradingAgents

Open-source multi-agent LLM trading framework where analysts, bull/bear researchers, a trader, and a risk manager debate before deciding.

by Tauric ResearchListed August 4, 2026

TradingAgents is an open-source Python framework that simulates the structure of a real trading firm. Seven roles — fundamentals, sentiment, news, and technical analysts, a researcher, a trader, and a risk manager — debate their views before a portfolio manager approves or rejects each transaction proposal, built on LangGraph.

It supports many LLM providers including OpenAI, Anthropic, Google, DeepSeek, and local models via Ollama, and trades on a simulated exchange. TradingAgents is a research and decision framework for building and studying AI trading systems, not a brokerage connection.

Tagged with

open-source multi-agent research

Related tools

View all

AI Hedge Fund

New

Open-source multi-agent stock analysis — up to 19 investor personas (Buffett, Burry, Wood, Druckenmiller) debate and backtest picks.

Signal Engines⚡ agentic

Equities

AI-Trader

New

Open-source, agent-native trading platform by HKUDS — agents register via SKILL.md, collaborate, and trade live across brokers and assets.

Autonomous Execution⚡ agentic

Equities · Options +3BinanceCoinbase+1 more

Astor

New

YC-backed personal AI investment analyst for retail — 24/7 market monitoring, real-time alerts, and research tailored to your portfolio.

Portfolio & Research

Equities