📊TradingAgents
TauricResearch/TradingAgents · homepage ↗
TradingAgents: Multi-Agents LLM Financial Trading Framework
⭐ Hugely popular: 98k stars, gaining about 1k a week
View on GitHub ↗repo profile
momentum
durability
bus factor = how many people it takes to cover more than half the commits (6 months). 1 is a solo project; higher means the work is spread across a team. top-author share is the single busiest author's slice of those commits.
since we covered it
why it's a big deal
- Packages the whole trading-analysis workflow, fundamentals, sentiment, news, technicals, research debate, execution, and risk sign-off, into one reproducible framework so researchers can study multi-agent decision-making instead of wiring it up themselves.
- Targets people testing LLMs on financial reasoning rather than live traders; the authors state it is for research only and not investment or trading advice.
- Splits the problem into distinct analyst, researcher, trader, and risk roles, letting bullish and bearish researchers debate before a portfolio manager approves or rejects a trade.
under the hood
- Built in Python and orchestrated with LangGraph, which handles the modular graph of agents and their message passing.
- Model-agnostic across OpenAI, Anthropic, Google, xAI, DeepSeek, Qwen, GLM, MiniMax, OpenRouter, Ollama, Azure, and any OpenAI-compatible endpoint, with separate deep-thinking and quick-thinking model slots.
- Exposes configuration for debate-round count, temperature for reproducibility, checkpoint resume for interrupted runs, and decision logging with memory persistence.
Radar summary, generated from the project's public sources
star history
- PR#10 2k 2025-06-11
- now 98k + 96k since first covered
curve is sampled from GitHub's star history, plus our own daily readings since we covered it; the dashed stretch is before we first covered it, the solid line since. figures at coverage are the numbers we printed then (approx.), current count is live.
understory
Better known than its recent output, coasting a little on attention.
- output, commits & releases
- clout, star velocity
output = commits & releases; clout = star velocity, both 0 to 100 monthly indices; the gap where output runs above clout is the understory. The understory →
covered in
-
Open multi-agent RL simulation for markets
similar projects
compare these →- 🦌 DeerFlow
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.
80k ACTIVE - 🗜️ headroom
Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
66k ACTIVE - 🧠 LangGraph
leaner, 40k stars
Build resilient agents.
40k ACTIVE
comments
Sign in with GitHub to add your blip on TradingAgents.