Analysis updated 2026-05-18
Study prediction-market trading strategies with simulated money instead of real funds
Compare Conservative, Moderate, and Aggressive mean-reversion trading approaches side by side
Watch live Polymarket signals and model performance on a web dashboard
Deploy a hosted, always-on version of the trading dashboard for around $7 a month
| thinkpixeliab/polymarket-ai-trading | alchaincyf/codex-orange-book | gainubi/note-slides | |
|---|---|---|---|
| Stars | 156 | 160 | 151 |
| Language | HTML | HTML | HTML |
| Setup difficulty | moderate | easy | easy |
| Complexity | 3/5 | 1/5 | 2/5 |
| Audience | vibe coder | vibe coder | vibe coder |
Figures from each repo's GitHub metadata at analysis time.
Requires Docker and an OpenAI API key for the AI-assisted scoring features.
Polymarket AI Trading System is a paper-trading research tool for Polymarket prediction markets, markets where people bet on the outcomes of real-world events. The emphasis on "paper trading" means all trading in this system is simulated: no real money moves by default, making it a safe environment to study and experiment with trading strategies. The system pulls live market data from Polymarket's order book API and runs three parallel simulated trading "personalities", Conservative, Moderate, and Aggressive, all using the same underlying mean-reversion strategy (the idea that market prices that have moved far from their historical average tend to drift back). The three models differ mainly in how aggressively they size positions and how strong a signal they require before acting. An optional OpenAI integration can provide AI-assisted analysis and market quality scoring. All activity is logged to a local SQLite database (a simple, file-based database), and a web dashboard shows live signals, model performance statistics, and AI analysis panels. The dashboard is hosted as a static site and talks to a backend server running as a Node.js application. Running the full system locally requires Docker. A hosted example dashboard and backend are publicly accessible. For always-on hosting, the README describes deploying to Render for approximately $7 per month. The full README is longer than what was provided.
A paper-trading dashboard that simulates three AI trading strategies against live Polymarket prediction market data, no real money involved.
Mainly HTML. The stack also includes Node.js, Express, SQLite.
License terms are not stated in the provided README excerpt.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly vibe coder.
This repo across BitVibe Labs
Verify against the repo before relying on details.