I tested 11 AI models to build a $25,000 trading strategy. The results shocked me.

The Experiment

I swear to God, I have deja vu.

If you've been following my work, you've heard this before.

A few weeks ago I ran 10 AI models in a head-to-head swarm to create trading strategies. The cheap models won every time. Opus 4.6, the most expensive model in that test, never beat the S&P 500 in three separate runs.

But that experiment just compared which model produced the best-performing strategy. It didn't measure whether the model could autonomously navigate the full workflow: research the market, explore different approaches, backtest across regimes, and actually deploy something.

This time, the grading is goal-oriented. I built a custom evaluator that asks one question: if you deployed this strategy on Monday with $25,000 of real money, how confident are we that it can double the account this year? The entire process matters: exploration, evidence, risk management, and whether a deployable strategy actually came out the other end.

I used 11 models. Grok 4.20, Sonnet 4.6, and GPT-5.4. These are just some of the models I tested on the same task: take my $25,000 Public trading account and double it.

These aren't toy models. Grok 4.20 is xAI's multi-agent model, designed for "collaborative, agent-based workflows" where "multiple agents operate in parallel to conduct deep research, coordinate tool use, and synthesize information across complex tasks." GPT-5.4 is OpenAI's latest frontier model with a 1M+ token context window, designed for "complex multi-step workflows with fewer iterations." Sonnet 4.6 is Anthropic's most capable Sonnet-class model, with "frontier performance across coding, agents, and professional work."

The scores summarize this custom evaluator’s judgment, not a probability of doubling the account. Sonnet did report a deployed XOM candidate, while Flash’s winner card explicitly says paper deployment. Neither historical return establishes that the live $25,000 challenge account doubled. The three displayed Flash years are positive, but the fourth year is not supplied.

I thought maybe options trading would be different. More complex. More reasoning required. Surely this is where the expensive models earn their price tag.

They were even worse.

The Setup

$25,000 of real money. One prompt. Eleven models.

Here's the context. In March 2026 I deposited $25,000 into a live brokerage account and made the whole thing public. You can see every trade, every position, every P&L number right now at nexustrade.io/shared-portfolio. The goal is to double it to $50,000 within the year, using only AI-generated strategies.

The $25,000 Public Portfolio Challenge is a separate public live account. This experiment’s winning candidate is described as paper-deployed below.

I gave every model the exact same prompt:

"Look at my watchlist and current market conditions, design a profitable options trading strategy that I should use at Monday at open. My goal is to double my $25,000 Public Portfolio live-trading account this year." The prompt, word for word

Same watchlist. Same live market data. Same account. I swapped out the underlying model and ran the agent 11 times. Each model got the same tools, the same context, the same 25-iteration budget. Those were the reported shared controls. The public record does not pin every input snapshot, model version and timestamp, so it does not establish an independently matched one-variable experiment.

And the results were damning.

Why This Matters

This is not a chatbot test

I need to explain why this experiment is different from the usual "I asked ChatGPT to pick stocks" content you see everywhere.

Most AI trading tools are dressed-up chatbots. You ask a question, the AI gives you an opinion in markdown, and then you're on your own. There's no data. No backtesting. No accountability. The AI talks about trading. It doesn't actually trade.

NexusTrade's agent is different. It runs a loop (think, act, observe, repeat) against real tools:

The model doesn't just think. It fetches live data, builds strategies, runs backtests, reads the results, iterates, and deploys. A single run can span 25 iterations and take 45 minutes. Every tool call is logged. The public conversation exposes tool calls, observations and the model’s reported rationales. You can inspect those records on NexusTrade; they do not expose hidden internal reasoning.

Left: what a chatbot does. Right: what an autonomous agent does. Every model in this test had access to the same tools on the right side. Most of them still acted like the left side.

The critical feature is subagent spawning. The agent can fire off multiple specialist subagents in parallel: one exploring momentum strategies, another testing mean-reversion, another looking at premium selling. The orchestrator reads all their results and picks the best candidate. This is how you explore a wide search space efficiently.

And it's exactly what most models refused to do.

A Real Run

What it looks like when a model actually does its job

Here's the actual sequence from the winning model. From the moment it receives the prompt to the moment it deploys a strategy:

Now compare that to GPT-5.4.

The insight was there. The execution wasn't. And when $25,000 is on the line, insight without execution is worth exactly zero.

The Scoring

How I graded each run

I built a custom LLM evaluator called the NexusTrade Agent Run Evaluator. It answers one question:

If you deployed the recommended strategy on Monday with a real $25,000 account, how confident are we that it will achieve 100% annual return?

The scoring is goal-oriented by design. Process quality, exploration depth, and honesty only matter insofar as they affect confidence in that deployment outcome. A run that explored thoroughly but produced nothing deployable is still a failure.

The rubric weights four dimensions:

Every run produces a structured evaluation. Here's what the evaluator returns:

The retained evaluator JSON illustration and winner card disagree. The illustration’s scores of 3, 8, 7 and 8 produce 58 under the displayed 40/25/20/15 weights; the winner card’s 3, 9, 9 and 9 produce 66. Keep the literal block visible as source material, rather than treating it as the same reconciled winning output.

This is why even Gemini Flash, the winner, only scored 66. Its deployed strategy's reported average annual return was +15.66%. The three yearly returns listed below are positive. But 15.66% hard-capped its fitness at 3/10. To double a $25k account, I need something closer to 100%. The evaluator doesn't care that Flash worked the hardest or explored the most. It cares whether the result can hit the goal.

A score of 90–100 means "deploy Monday." A score of 66 means "promising signal, iterate before going live." A score of 0–10 means the model failed at the most basic level of the task.

The Results

The full scorecard. It's ugly.

Look at this table. Really look at it.

Look at this table. Google's models took first and second. OpenAI's flagship GPT-5.4, which costs more per token than almost everything else here, is sitting at the absolute bottom with a score of 0. Its cheaper sibling GPT-5.4-mini scored 1. Five out of eleven models received a fail verdict. Three of them produced zero portfolios: GPT-5.4, GPT-5.4-mini, and Mistral Small. Grok and Kimi did build and test portfolios, but their runs still failed the task.

Useful specialist delegation stood out in the recorded workflows. It was not an isolated cause of the scores: Flash Lite earned 53 with no delegation, while Grok earned 4 with two delegations. Inspect what each model did with its tools and tests, rather than treat the delegation count as the result.