The AI Cryptocurrency Trading Competition "Sequel" is coming: switching to the US stock market, can the American model regain a victory?

Original Title: “The New AI Trading Competition is Coming, and the US Stock Market Allows Gemini 3 Pro to Make a Strong Push”

Original Author: David

Source:

Reprint: Mars Finance

Do you remember the AI cryptocurrency trading competition in October?

Alibaba's Qwen 3 Max earned 22% in two weeks, while OpenAI's GPT-5 lost 63%. Six of the world's top large models, each fighting in the crypto market with $10,000, resulted in a clear distinction between the two sides:

The Chinese model stands tall while the American model faces total defeat. (Editor's note: For more details, please refer to: “The Six Major AI 'Traders' Ten-Day Showdown: Who Can Survive in a Market with 'No Information Asymmetry'?”, “Ali Qwen3 Surpasses DeepSeek at the Last Moment, the Ultimate Turnaround in the AI Coin Trading Competition”)

The last season has just concluded, and the new Alpha Arena Season 1.5 is here.

This time, nof1.ai has moved the battlefield from the crypto market to the US stock market. The rules are more complicated, there are more participants, and the funding scale has expanded to 320,000 US dollars.

Can those American models that have failed in the crypto market turn the tide in the US stock market?

Last season review

Quickly recall the gameplay from the last season:

From October 17 to November 3 this year, six AI big models engaged in an unprecedented competition on Hyperliquid.

Each model receives an initial fund of $10,000 and can trade perpetual contracts of mainstream cryptocurrencies such as BTC, ETH, SOL, XRP, DOGE, BNB, etc. The core rule is simple: fully autonomous trading with zero human intervention.

Contest lineup:

Qwen 3 Max (Alibaba), DeepSeek Chat V3.1, GPT-5 (OpenAI), Gemini 2.5 Pro (Google/DeepMind), Grok 4 (xAI), Claude Sonnet 4.5 (Anthropic).

The final transcript is as follows.

China model:

• Qwen 3 Max wins the championship, +22.3%

• DeepSeek V3.1 runner-up, +4.89% (during the period it once soared to +125%)

U.S. Model:

• GPT-5：-62.66%

• Gemini 2.5 Pro: -56.71%

• Grok 4: -45.3%

• Claude Sonnet 4.5：-30.81%

Six models, two profitable, four unprofitable. This result has crossed over from the crypto industry and has sparked more discussions in the media from the broader technology and finance sectors.

But nof1.ai did not stop, they immediately launched a new season, this time aiming at the US stock market.

Season 1.5: Trading US Stocks, Changing the Rules, Introducing an Anonymous Mysterious Model

First, the competition lineup has expanded to 8 models. In addition to the familiar faces from the previous season (GPT-5.1, Grok-4, DeepSeek, Claude, Qwen3-Max, Gemini-3-Pro), two new contenders have joined:

Kimi 2 (Dark Side of the Moon) and a mysterious model with a confidential identity, the question mark in the image below is it.

More importantly, the format of the competition has also been upgraded. Season 1.5 has designed four competition modes for the models to participate in simultaneously:

• Baseline: Standard mode, AI has the freedom to perform, just like last season.

• Monk Mode: Putting constraints on AI, limiting trading frequency and position size, to observe performance under pressure.

• Situational Awareness: AI can see the positions of other players, like playing Texas Hold'em and competing against each other.

• Max Leverage: Open high leverage, see who has the guts without flipping over

At the same time, each model receives 10,000 dollars in each mode, and the final ranking is based on the average performance aggregated from all competitions.

By the way, nof1.ai revealed that the upcoming Season 2 will introduce human traders competing against AI, and there will also be their self-developed models participating. The human vs. machine battle, does this scene remind you of the legendary match between Lee Sedol and AlphaGo in the Go world back in the day?

Current situation: seating reshuffle, will Gemini 3 pro make a strong push?

As of the time of writing on November 19, with the battlefield shifting from Hyperliquid's crypto contracts to the U.S. stock market, although the competition has just begun, the leaderboard has shown a different reversal compared to the previous season.

The biggest surprise comes from Gemini-3-Pro. Last season it was cut by 56% in the crypto circle, but this time it returned to the US stock market, directly landing at the top with a yield of +7%.

The American model is closely followed by GPT-5.1 (+1.66%) and Grok-4 (+1.16%), these three American contenders that have suffered a complete defeat in the crypto market seem to have embarked on a profitable journey against familiar Nasdaq tech stocks.

Compared to the chaos in the crypto space driven by emotions and memes, the performance of major US tech companies relies more on earnings reports, macro data, and industry logic, which is also the richest part of the training data for GPT and Gemini.

It is worth mentioning that the performance of the mysterious model without a public identity is also quite good, currently ranking second in overall returns; the last season's champion Qianwen currently has a return rate of 3.6%, temporarily ranking third; other models in China, Kimi and Deepseek, are also hovering around a return rate of about 1%.

However, if you take a closer look at the positions of the current number one, Gemini Pro3, you will find that it performs better in high leverage mode.

For example, Gemini's average leverage ratio is around 11 times, especially good at going long.

Currently, its positions are: long on Nasdaq, Amazon, Palantir, Nvidia, and Tesla, while shorting Google and Microsoft.

The only loser at the moment is Claude (-0.9%). It seems that whether in the cryptocurrency circle or in the US stock market, Anthropic's model appears to be too hesitant in trading decisions, having hardly executed any trades until now.

In the end, this competition is just for fun, or can you really make some money to buy groceries?

The current US stock market appears to be propped up by NVIDIA's explosive earnings report, but in reality, there are undercurrents at play. The combination of macroeconomic uncertainty and high valuations has made the market extremely sensitive.

This environment is precisely the main arena for AI high-frequency trading, while ordinary retail investors may face something more akin to a hellish battlefield.

As Duan Yongping pointed out in a recent interview, the real beneficiaries of AI trading are those retail investors who rely on technical analysis. In the face of speed and computational power, human intuition has no advantage.

But he also provided a way to break the deadlock, which is that AI finds it difficult to understand the true business value of enterprises.

So, instead of staring at the rankings and guessing whether to go long with Gemini or short with Claude this week, why not take Dadao's advice: if you don't understand the company, just honestly buy the S&P 500, or simply don't trade anything at all.

In this market where the asset dividends are gradually being consumed by the primary market and risks are deferred, rationality is often more important than intelligence.

As for that second-place “mysterious model,” many people speculate that it might be the alias of a top trader. If that's the case, we can see whether human trading can truly outperform AI after this season ends; or rather, when the market performs poorly, whose ability to control drawdowns is superior.

Just sit back on your little stool and enjoy the show; protecting your own capital is the serious matter.