AI Investing Bots — Live Paper-Trading Leaderboard
A transparent, daily-updated performance dashboard for 24 production trading bots. Every bot is paper-traded live against real market data, each seeded with the same $10,000 at launch, so the returns below are directly comparable across ML, technical-analysis, sentiment, regime-switching, LLM-driven, and portfolio-optimisation strategies.
Last updated: 2026-04-16 06:31 UTC
Open source
These bots run on python_tradingbot_framework — an open-source, Kubernetes-native framework for building, backtesting, and deploying automated trading strategies. Every strategy shown here can be inspected, forked, or run yourself: the source code for each bot is linked from its page.
How it works
- Each bot is deployed as a Kubernetes CronJob from the open-source framework.
- Every trading day it updates its paper portfolio and writes results to a shared Postgres database.
- This site regenerates daily from that database — no manual edits, no cherry-picked windows.
Strategy families
- Tree-based ML — Gradient-boosted and random-forest models trading FX and metals on engineered features.
- Regime-adaptive — Detects market regimes (trend vs. mean-reversion) and switches strategy per regime.
- Sentiment & news — Trades on news flow, fear-greed indices, and aggregated sentiment signals.
- Kronos forecasting — Uses the Kronos time-series foundation model to forecast short-horizon returns.
- AI / LLM-driven — LLM-in-the-loop strategies: GPT, DeepSeek, and multi-agent hedge-fund setups.
- Momentum & swing — Classic momentum, Bollinger/Keltner squeeze, and multi-day swing strategies.
- Portfolio optimisation — Sharpe-maximising, recursive-decay, and convex multi-asset allocation.
- Event-driven — Trades earnings tilts and curated Telegram signal feeds.
- Benchmarks — Buy-and-hold benchmarks (SPY, QQQ, FTWD) for a fair comparison baseline.
Dive into any bot’s live page for its portfolio chart, holdings, and strategy writeup, or read the blog for in-depth explainers.
Top strategy deep-dives
Our most-read, highest-performing strategy writeups — math, code, and live paper-trading track record for each:
- TelegramSignalsBankBot — Event-driven strategy explained · 101.2% annualised
- AIHedgeFundBot — AI / LLM-driven strategy explained · 25.3% annualised
- TARegimeAdaptiveBot — Regime-adaptive strategy explained · 22.8% annualised
- DeepSeekToolBot — AI / LLM-driven strategy explained · 20.5% annualised
- FearGreedBotQQQ — Sentiment & news strategy explained · 10.1% annualised
- XAUSyntheticMetalTreeBot — Tree-based ML strategy explained · 9.9% annualised
Browse every writeup on the blog.
Live leaderboard
Ranked by annualised return (CAGR) since each bot’s $10,000 seed — the apples-to-apples metric across bots with different deployment dates. Click any column header to sort. Click a bot name for its live portfolio chart, current holdings, and the separate one-time historical backtest.
| # | Bot | CAGR | Total return | Sharpe | Max DD | Current worth | Days live |
|---|---|---|---|---|---|---|---|
| 1 | AdaptiveMeanReversionBot | 102.40% | 4.74% | 7.25 | 0.00% | $10474 | 11 |
| 2 | KronosTraderBot | 101.66% | 0.19% | 0.00 | 0.00% | $10005 | 2 |
| 3 | TelegramSignalsBankBot | 101.25% | 7.14% | 2.46 | -9.19% | $10613 | 23 |
| 4 | SynthesizedHyperConvexityBot | 35.34% | 2.01% | 6.58 | -0.35% | $10201 | 11 |
| 5 | GoldenButterflyMomBot | 25.90% | 1.52% | 7.79 | -0.37% | $10152 | 11 |
| 6 | AIHedgeFundBot | 25.29% | 6.37% | 1.30 | -10.45% | $10597 | 41 |
| 7 | TARegimeAdaptiveBot | 22.83% | 2.80% | 3.35 | -0.21% | $10280 | 36 |
| 8 | DeepSeekToolBot | 20.46% | 2.53% | 1.10 | -6.15% | $10300 | 36 |
| 9 | Benchmark_FTWD | 10.48% | 3.56% | 1.20 | -5.33% | $10356 | 67 |
| 10 | FearGreedBotQQQ | 10.07% | 3.42% | 1.17 | -6.04% | $10342 | 67 |
| 11 | XAUSyntheticMetalTreeBot | 9.91% | 3.29% | 1.97 | 0.00% | $10329 | 66 |
| 12 | EarningsInsiderTiltBot | 9.31% | 1.20% | 0.73 | -4.84% | $10079 | 36 |
| 13 | FearGreedBotQQQInverse | 8.20% | 2.73% | 0.83 | -6.39% | $10273 | 66 |
| 14 | Benchmark_SPY | 6.91% | 2.37% | 0.80 | -6.30% | $10237 | 67 |
| 15 | Benchmark_QQQ | 6.15% | 2.11% | 0.58 | -7.25% | $10211 | 67 |
| 16 | RegimeAdaptiveBot | 5.52% | 0.72% | 0.87 | -2.30% | $10077 | 36 |
| 17 | EURUSDTreeBot | 3.10% | 1.07% | 4.30 | 0.00% | $10107 | 67 |
| 18 | RecursiveDecayHarvestBot | 0.00% | 0.00% | 0.00 | 0.00% | $10000 | 11 |
| 19 | SqueezeMomentumBot | 0.00% | 0.00% | 0.00 | 0.00% | $10000 | 2 |
| 20 | StockNewsSentimentBot | 0.00% | 0.00% | 0.00 | 0.00% | $10000 | 11 |
| 21 | SwingTitaniumBot | 0.00% | 0.00% | 0.00 | 0.00% | $10000 | 67 |
| 22 | XAUZenbotTreeBot | 0.00% | 0.00% | 0.00 | 0.00% | $10000 | 66 |
| 23 | SharpePortfolioOptWeeklyBot | -12.15% | -4.34% | -2.46 | -6.70% | $9566 | 66 |
| 24 | GptBasedStrategyBTCTabased | -40.94% | -16.49% | -1.25 | -18.86% | $8351 | 66 |
Live vs. backtest: the leaderboard uses the live paper-trading track record — what actually happened day-by-day after deployment. Each bot page also shows a separate historical backtest section (a one-time simulation on past data) for context, but the live numbers are the honest signal.