No-code trading bot results: the first full-window numbers

A no-code trading bot keeps its own scoreboard, and Finforge prints every row of it. In its Sophon-3 paper, Version 1.0, published in July 2026, the three forecasting agents show full-window compound growth of 105.2% for Apex, 62.2% for Core and 77.6% for Surge against index benchmarks of 30.9% and 20.9%. The paper tells you how to read these numbers before it shows them. These are simulated full-window results from Finforge's research, not an account statement, and past performance is not a guide to future returns.

The scoreboard sits in the empirical results section of the paper. Each of the three agents runs against a price index that serves as its benchmark, the Nasdaq-100 for Apex and Core and the S&P 500 for Surge. The window runs from early 2023 to 1 July 2026.

What a no-code trading bot's scoreboard shows

Apex, on the Nasdaq-100, shows a compound annual growth rate of 105.2% against the index's 30.9%. CAGR is the smoothed yearly growth figure. Core, also on the Nasdaq-100, shows 62.2% against the same 30.9%. Surge, on the S&P 500, shows 77.6% against that index's 20.9%. Every agent beat its benchmark on these full-window figures.

The paper also prints a return-for-risk score, called Sharpe, with the risk-free rate set to zero. Higher is better. Apex's Sharpe is 2.02 against the index's 1.43, Core's is 1.80 against 1.43 and Surge's is 1.80 against 1.35. Each agent carries a maximum drawdown, the deepest peak-to-trough loss on its own path. Apex's is -29.5%, Core's -23.8% and Surge's -24.7%.

Active drawdown is the one worth watching. It measures the deepest underperformance of the portfolio against its benchmark, how far the relative path dips below it. Apex's active drawdown is -28.7%, Core's -18.3% and Surge's -18.7%. A bot can fall with its market. The question is how far it falls behind it.

Why the paper tells you not to trust the headline

The paper leads with the caveat. These full-window tables span the configuration boundary and the architecture freeze, two design milestones, so they are not clean design-frozen out-of-sample evidence. The clean slice is reported separately. Read the headline figures as mixed, not settled.

The live account is real but young, at roughly 41 trading days, and the paper says that is not statistically significant on its own. A short record cannot prove an edge, however good it looks.

What a no-code trading bot cannot claim

The zero-slippage base case is a property of the regime, not a universal claim. These agents place small orders in highly liquid large-caps at the close, so at this size there is no market impact. The paper is silent on capacity. It says nothing here is evidence of institutional capacity.

The deep component's demonstrated value, the paper says, is forecast-magnitude sizing, not directional forecasting. That is the finding that tends to surprise people who assume the AI's job is picking direction. The formalism is topology-general, but the empirical evidence covers one topology.

Questions people ask

Which no-code trading bot returned the most over the full window?

Sophon Apex shows the largest full-window figure, 105.2% compound annual growth rate against the Nasdaq-100's 30.9%, from early 2023 to 1 July 2026. Core shows 62.2% and Surge shows 77.6%. The paper says to hold these lightly, because the window spans two design milestones and is not clean out-of-sample evidence. Past performance is not a guide to future returns.

Is Finforge a no-code bot builder?

No. Finforge is not a toolkit where you assemble and test your own strategies. It ships pre-built live agents, Sophon Apex, Core, Pillar and Surge, trading founder capital in Alpaca paper accounts in public since 21 April 2026. The no-code promise is that the quant machinery runs for you, not that you build it. No customer money is traded before launch.

Finforge runs its Sophon agents in public and publishes every trade, benchmark and drawdown. Check the live results, benchmarks and drawdowns for every Sophon agent at finforge.com/agents, read the full method in the Sophon-3 research paper, and join the waitlist.