📊 Full opportunity report: Introducing Forezai · TradingAgents — a committee of LLMs decides paper-trades on ThorstenMeyerAI.com — validation score, market gap, and execution plan.

TL;DR

Forezai has launched TradingAgents, a framework where multiple specialized large language models (LLMs) collaborate to make paper-trading decisions. This development aims to explore whether AI committees can outperform random choices in trading simulations, marking a step forward in AI-driven financial research.

Forezai has launched TradingAgents, a fork of an existing multi-agent LLM framework, adding operational features to enable autonomous paper trading based on collaborative AI decision-making. This development aims to test whether a committee of specialized large language models can produce trading decisions that are at least no worse than random chance, advancing research in AI-driven trading strategies.

The original framework, developed by TauricResearch, involves thirteen agent roles that analyze market data from multiple perspectives, including fundamentals, news, social media, and market structure. These agents debate and synthesize their findings through structured stages, culminating in a final trading recommendation. The new Forezai fork incorporates an operational layer, enabling automated daily execution of paper trades, position management, and comprehensive logging, all within a local, cloud-free environment.

Key features of the Forezai system include an autonomous scheduler, a multi-broker abstraction supporting local, Alpaca paper trading, and a web dashboard for real-time monitoring. The setup is designed for research, not live trading, with multiple safeguards to prevent accidental real-money exposure. The system is intended to evaluate whether a committee of LLMs can generate trading signals that outperform random decisions, with the process emphasizing explicit reasoning over mere prediction.

Introducing Forezai · TradingAgents — Thorsten Meyer AI
AGENTS
● ANNOUNCEMENT / MAY 2026
THORSTEN MEYER AI · FOREZAI · § 03
FOREZAI · 03
TRADINGAGENTS · LAUNCH
Research Series · Companion to Polybot Week 1-2 · 2026-05-17

Introducing Forezai · TradingAgents.
A committee of LLMs
decides paper-trades.

After two weeks of finding out most parametric strategies don’t work, the obvious next research question: can multi-agent LLM judgment do any better?
A fork of the open-source TradingAgents framework (TauricResearch): thirteen LLM agents in four stages — four parallel analysts · a bull-bear debate with research-manager arbitration · a three-voice risk team · a two-layer trader + portfolio-manager decision. The fork keeps the agent graph intact and adds the operational layer the upstream doesn’t ship: an autonomous loop · a multi-broker abstraction · a local web dashboard · Codex OAuth · MCP plug-ins · 520+ unit tests. The question is narrower than “do LLMs predict the market” — that prior is “no, with high confidence.” The narrower question is: when LLMs are structured into specialised adversarial roles, does the committee produce decisions at least no worse than a coin flip after fees? Honest priors before running: it might fail too. If it appears to work, the most likely explanation is variance.
This is not financial advice. Nothing in this announcement should be used to inform real trading decisions. The software described trades simulated money by default. If you reconfigure it to trade real money, you should expect to lose that money — regardless of how clever any individual agent’s reasoning looks. Algorithmic trading is zero-sum after fees and structurally hostile to part-time retail strategies.
13 agents
Specialised roles in four stages
Analysts · Debate · Risk · Decision
78% / -33%
Polybot prior: fleet win rate
combined with -33% bankroll
520+
Passing unit tests across engine,
services, HTTP routes (starting baseline)
€0 floor
LLM cost on Codex OAuth
(falls back to public API per token)
FOREZAI / TRADINGAGENTS· APACHE 2.0 FORK· UPSTREAM TAURIC RESEARCH· LANGGRAPH· 13 AGENTS / 4 STAGES· 4 PARALLEL ANALYSTS· BULL-BEAR DEBATE· 3-VOICE RISK TEAM· TRADER + PORTFOLIO MANAGER· 5-TIER FINAL RATING· ALPACA PAPER + LOCAL + SHADOW· LIVE ENDPOINTS HARD-REFUSED· FASTAPI + REACT VIA CDN· CODEX OAUTH· MCP PLUG-IN REGISTRY· 520+ UNIT TESTS· POLYBOT WEEK 1: 21 EXPERIMENTS· WEEK 2: -33% BANKROLL· 78% FLEET WIN RATE· HONEST RESEARCH, NOT EDGE· FOREZAI / TRADINGAGENTS· APACHE 2.0 FORK· UPSTREAM TAURIC RESEARCH· LANGGRAPH· 13 AGENTS / 4 STAGES· 4 PARALLEL ANALYSTS· BULL-BEAR DEBATE· 3-VOICE RISK TEAM· TRADER + PORTFOLIO MANAGER· 5-TIER FINAL RATING· ALPACA PAPER + LOCAL + SHADOW· LIVE ENDPOINTS HARD-REFUSED· FASTAPI + REACT VIA CDN· CODEX OAUTH· MCP PLUG-IN REGISTRY· 520+ UNIT TESTS· POLYBOT WEEK 1: 21 EXPERIMENTS· WEEK 2: -33% BANKROLL· 78% FLEET WIN RATE· HONEST RESEARCH, NOT EDGE·
FIG. 01 — THE 13-AGENT COMMITTEE
Thirteen specialised roles · four stages · biases made to argue in public
The architecture forces the system to articulate its reasoning rather than relying on what a single context window happens to recall
Stage 1 · Four analysts in parallel4 agents
Market
Structure, ranges, regime indicators
News + Insider
News flow, filings, insider activity
Fundamentals
Balance sheet, earnings, ratios
Social Sentiment
Social-media tone, retail signal
Stage 2 · Bull-bear debate + research-manager arbitration3 agents
Bull researcher
Argues upside thesis from analyst reports
Bear researcher
Argues downside thesis from same reports
Research manager
Arbitrates · writes single synthesis
Stage 3 · Three-voice risk team3 agents
Aggressive
Looks for upside · accepts variance
Conservative
Looks for downside · protects capital
Neutral
Balances · forces downside articulation
Stage 4 · Two-layer decision2 agents
Trader
Three-tier proposal · buy / hold / sell
Portfolio manager
Five-tier rating + price target + horizon · sees arguments only, never raw data
The portfolio manager only sees the arguments, never the raw data — which forces the committee to make its reasoning explicit rather than relying on a single context window’s recall. The upstream framework ships the agent graph; it does not ship the operational machinery to run that graph on autopilot, observe its results honestly, store them for later inspection, or prevent the operator from accidentally trading real money. That gap is what the Forezai fork fills.
FIG. 02 — THE POLYBOT PRIOR · WHY THIS IS A DIFFERENT BET
Two weeks of paper-trading prediction markets · the trap underneath the headline numbers
25 experiments · 78% fleet-wide win rate · -33% bankroll · most parametric strategies are structurally negative-expectation when measured honestly
The flattering number
78%
Fleet-wide win rate · week 2
“You can win four out of five trades and still go broke, because the one loss is bigger than the four wins put together.” Win rate without P&L context is a mechanical illusion.
The honest number
−33%
Fleet bankroll · week 2 close
The strongest possible demonstration of the trap. A parametric trading strategy that looks compelling in a backtest will almost always fail to survive a fresh sample. Most “edges” are mechanical artefacts.
Week 1: 21 parallel strategy experiments · early winners mostly mechanical illusions · exactly one strategy (a fair-value taker on BTC) showed the mathematical signature of real edge over a few hundred settled trades. Week 2: same fair-value strategy with more data collapsed. A separate mid-week hypothesis (market-making) also failed cleanly. Fleet ended week 2 at roughly negative thirty-three percent of bankroll. The honest research finding wasn’t on the winning side — it was on the losing side. Adding more parameters to Polybot wouldn’t change that. TradingAgents is asking a separable question.
FIG. 03 — WHAT THE FORK ADDS · THE OPERATIONAL LAYER
Six layers the upstream framework doesn’t ship
Same agent graph, intact. The fork makes it a research instrument rather than a tech demo.
01 · Loop
An autonomous loop
Scheduler · watchlist · auto-trader maps ratings to paper orders · allow-list filtering · per-ticker cooldowns · sector caps · cash checks · position manager evaluates open positions every 60s for TP / SL / max-hold. Append-only audit logs.
02 · Brokers
Multi-broker abstraction
Three modes: local Python broker (yfinance fills, JSON-persisted) · Alpaca paper-trading adapter · “shadow” mode running both in parallel with divergence view. Real Alpaca live endpoints are hard-refused at multiple layers.
03 · Dashboard
A local web dashboard
FastAPI backend · React via CDN, no Node toolchain · SVG equity curve · rolling-peak drawdown · win-rate by rating / ticker / model · exit-reason breakdown · LLM cost vs realised P&L joined by run ID. Runs locally; nothing sent to a cloud service.
04 · Codex
Codex OAuth
Runs the engine on a ChatGPT Pro subscription via the Codex backend. LLM cost floor effectively zero if you already have ChatGPT Pro. Token stored encrypted locally. Falls back to the regular OpenAI API if you’d rather pay per token.
05 · Alerts
Multi-channel alerts
Slack · Discord · SMTP email · configurable filter on rating events and order fills · append-only history kept locally. Webhook URLs masked in API responses so a screenshot can’t accidentally leak credentials.
06 · MCP
MCP plug-ins
Registry for adding Anthropic Model Context Protocol servers (Kensho · Aiera · FactSet · Morningstar · LSEG) as analyst tools. Plug-ins advertise category (fundamentals · news · market data · social) · probe endpoint tests credentials.
Honest-by-design touches: every generated report prepends “Research, not advice” and appends a footer with version, commit, provider, models used, run ID, and cost. Closed trades carry the same metadata. 520+ passing unit tests across engine, services, and HTTP routes. The intent: when the system loses money, the journal makes it impossible to pretend it didn’t.
FIG. 04 — HONEST PRIORS · BEFORE RUNNING THIS IN ANGER
Three priors stated before the data starts arriving
The bias of the project: when the data says no, the dashboard says no, the article says no
1
It might fail too. LLMs are not oracles, and a sophisticated framework around language-model outputs does not change the underlying error rate of the model. Sample is still everything. The framework’s outputs are subject to the same statistical noise as any prediction system over small samples.
Highest likelihood
2
If it appears to work, the most likely explanation is variance. The same trap that caught the first article’s candidate edge applies here. A high win rate over fifty trades means much less than it looks. Without out-of-sample confirmation, a flattering early sample tells you almost nothing about whether the system has real edge.
Second-most likely
3
If it appears to work for the right reasons — empirical win rate matches stated confidence, and alpha-versus-benchmark persists across non-overlapping samples — that would be a meaningful research finding. Whether that happens, I don’t know. The point of putting it in the open is that the data will say.
Genuinely open
This is explicitly not a launch announcement for a product anyone should connect a real brokerage account to. The Alpaca live endpoints are hard-refused at multiple layers in the code, and the design choice is deliberate. The right next step is data, not deployment. The bias of the whole project is straightforward: when the data says no, the dashboard says no, the article says no, and no one tries to retroactively rescue the thesis. That’s the contribution.
FIG. 05 — WEEK THREE · WHAT THE METHODOLOGY WILL MEASURE
Four concrete measurements before publishing findings
The hope: write the week-three article from a position of “here’s what the data says”. The fear: another candidate falsified at higher sample. Both outcomes are publishable.
M1 · Sample discipline
Small watchlist for a few weeks before publishing
A handful of tickers across two or three sectors. Long enough to gather sample, narrow enough to keep attention on what’s actually happening per agent. Avoid the noise of a 65-ticker autonomous loop until the smaller version has been read carefully.
M2 · Calibration view
Stated confidence vs. realised win rate
When the system says “75% confident”, do the trades actually win 75% of the time? Same measurement applied to Polybot’s fair-value model. If the model is systematically over-confident, that bias dominates everything downstream.
M3 · Cost accounting
Cost per ticker · per rating · per profitable trade
With Codex OAuth the marginal LLM cost is effectively zero. With the public OpenAI API, each run is hundreds of agent turns. The honest question: does this scale economically if you ever did run it at real cost?
M4 · Non-overlapping windows
Alpha vs benchmark · out-of-sample
Not within-sample alpha — trivially inflatable. Hold out one period entirely, run the system on the next, then check whether the held-out result matches the in-sample stats. If they diverge sharply, the in-sample was curve-fit.
Open under Apache-2.0 with upstream cited from every relevant surface. Not open: the operator’s running results, the specific watchlist, the per-agent prompt customisations, the alert channels, the trade journals — kept local for the same reason Polybot’s per-experiment data is kept local. Publishing exact configurations encourages people to copy them with real money, which is the opposite of what an honest research project should do. Summary findings will be published. Recipes will not.
The bet is on a different mechanism, not a different parameter setting. The point is not to find a money-printing AI. The point is to put honest measurements of these systems into the public record — so the next person looking at the space starts a step further along than the last.
Thorsten Meyer AI · Introducing Forezai · TradingAgents · § 03

Potential Impact of AI-Driven Trading Committees

This development matters because it explores whether AI systems composed of multiple specialized models can collaboratively generate trading decisions that are more reliable than random guesses. If successful, it could influence future AI research in finance, pushing beyond traditional parametric strategies, which have shown limited robustness. The project exemplifies a shift toward transparent, multi-agent reasoning in trading, emphasizing explicit articulation of decision rationale rather than relying solely on predictive accuracy.

3-Tier Foldable Paper Display Stand - Portable Adjustable Showcase for Trade Shows, Cosmetics & Handbags, Retail Merchandising Platform for Boutique, Exhibitions (L197 in * W31.5 in * H46 in)

3-Tier Foldable Paper Display Stand – Portable Adjustable Showcase for Trade Shows, Cosmetics & Handbags, Retail Merchandising Platform for Boutique, Exhibitions (L197 in * W31.5 in * H46 in)

Certified Kraft Paper & Honeycomb Bionic Technology deliver exceptional structural integrity – ensuring even load distribution, sustained operational…

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Background on AI in Trading and Research Progress

Previous research by Thorsten Meyer and TauricResearch involved testing parametric trading strategies against prediction markets, revealing that most explicit-rule strategies fail to survive real-market conditions despite promising backtests. These findings underscore the difficulty of developing robust, rule-based algorithms for trading. In response, researchers have turned to AI methods, particularly large language models, to see if their collective reasoning can improve decision-making in complex, uncertain environments. The TradingAgents framework was introduced as a way to structure multi-agent debate and synthesis, but until now, it lacked operational features for autonomous testing.

Forezai’s fork addresses this gap by integrating operational components, enabling continuous, automated paper trading and detailed performance logging, thus transforming the prototype into a practical research instrument.

“The goal is to see if a committee of LLMs, each with different biases and roles, can produce decisions that hold up against randomness, moving beyond simple prediction models.”

— Thorsten Meyer

Python for Algorithmic Trading: From Idea to Cloud Deployment

Python for Algorithmic Trading: From Idea to Cloud Deployment

New Store Stock

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Uncertainties About AI Trading Committee Effectiveness

It remains unclear whether the LLM committee will outperform random decision-making in live or simulated trading environments. The system’s effectiveness depends on how well the models can articulate and synthesize reasoning under real market conditions, which is still under testing. Additionally, the impact of model biases and the robustness of the decision process across different market scenarios are not yet fully understood.

AI Trading for Beginners: Master Stocks, Crypto, Options, and Forex with ChatGPT Prompts, AI Tools, Risk Management, and Proven Strategies

AI Trading for Beginners: Master Stocks, Crypto, Options, and Forex with ChatGPT Prompts, AI Tools, Risk Management, and Proven Strategies

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Next Steps for Testing and Validation

The project will conduct ongoing experiments using the Forezai system across various market conditions, collecting data on decision quality and performance metrics. Researchers plan to analyze whether the LLM committee’s decisions can consistently match or outperform baseline random strategies in paper trading. Future updates may include refining the agent roles, expanding the data sources, and eventually testing the framework with live trading under strict safeguards.

Advanced Stock Market Analysis Dashboard: Design and Implementation of a Real-Time Financial Analytics Platform

Advanced Stock Market Analysis Dashboard: Design and Implementation of a Real-Time Financial Analytics Platform

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Key Questions

Can Forezai’s TradingAgents system trade with real money?

No, the current implementation is designed for paper trading only, with safeguards to prevent accidental real-money trading. Transitioning to live trading requires deliberate override and risk assessment.

How does the multi-agent system improve decision-making?

It encourages explicit reasoning by having specialized models debate and synthesize their insights, aiming to produce more balanced and transparent trading decisions than single-model predictions.

Will this system outperform traditional trading algorithms?

It is currently experimental. The primary goal is to assess whether a committee of LLMs can produce decisions at least comparable to random chance, with future improvements potentially enhancing performance.

What are the risks of using AI for trading research?

AI systems can produce biased or inconsistent decisions, especially in volatile markets. The current setup is for research and simulation, not live trading, to mitigate these risks.

Source: ThorstenMeyerAI.com

You May Also Like

Mistral. The fourth path.

Mistral raises $830M in 2026, becoming Europe’s leading venture-backed AI firm, but faces limitations compared to US frontiers on top-tier capabilities.

The Skills Marketplace, Six Months Later: Predicted vs Actual

An analysis of the emergent skills marketplace six months after predictions, highlighting confirmed developments, structural surprises, and future outlook.

Forward-Deployed Engineer Economics 2.0: The Unit Economics Math, Six Months Later

Six months after initial analysis, the economics of FDE roles reveal profitability depends on contract size and customer cohort, shaping AI enterprise scaling.

Data processing agreement tracker for micro SaaS teams

A new data processing agreement tracker tailored for micro SaaS teams is being tested, aiming to streamline vendor and customer data compliance workflows.