Bitget App
Trade smarter
Buy cryptoMarketsTradeFuturesEarnWeb3SquareMore
Trade
Spot
Buy and sell crypto with ease
Margin
Amplify your capital and maximize fund efficiency
Onchain
Going Onchain, without going Onchain!
Convert & block trade
Convert crypto with one click and zero fees
Explore
Launchhub
Gain the edge early and start winning
Copy
Copy elite trader with one click
Bots
Simple, fast, and reliable AI trading bot
Trade
USDT-M Futures
Futures settled in USDT
USDC-M Futures
Futures settled in USDC
Coin-M Futures
Futures settled in cryptocurrencies
Explore
Futures guide
A beginner-to-advanced journey in futures trading
Futures promotions
Generous rewards await
Overview
A variety of products to grow your assets
Simple Earn
Deposit and withdraw anytime to earn flexible returns with zero risk
On-chain Earn
Earn profits daily without risking principal
Structured Earn
Robust financial innovation to navigate market swings
VIP and Wealth Management
Premium services for smart wealth management
Loans
Flexible borrowing with high fund security
Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead.

Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead.

ForesightNews 速递ForesightNews 速递2025/10/20 09:43
Show original
By:ForesightNews 速递

Deepseek currently ranks first, while Gemini is at the bottom.

Deepseek currently ranks first, while Gemini is at the bottom.


Written by: 1912212.eth, Foresight News


In the world of cryptocurrency, human traders are often troubled by emotions and information asymmetry. But what would happen if AI models were in charge of trading?


On October 18, a project called Nof1 placed several AI models—GPT-5, Claude Sonnet 4.5, Gemini 2.5 Pro, Deepseek V3.1, and Qwen3 Max—into the real crypto market, allowing them to autonomously make long and short trades on BTC, ETH, SOL, BNB, DOGE, and XRP assets on Hyperliquid.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 0


Nof1 is not a simple simulation, but a real-money battle: each AI model starts with $10,000, aiming to maximize profits in the highly volatile crypto market through intelligent algorithms.


Nof1's official website (nof1.ai) clearly displays real-time price charts and account value curves. Interestingly, the official team has also added a BTC Holder section for comparison, whose strategy is simply to buy and hold BTC.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 1



As of 11:00 am on October 20, the total account values of the major models fluctuate above $10,000. DeepSeek, founded by Liang Wenfeng and backed by a Chinese quantitative fund, currently ranks first with a position value of about $11,800. Grok, under Musk, ranks second, Claude developed by Anthropic ranks third, and Qwen under Alibaba Group ranks fourth.


Most surprisingly, OpenAI's latest large model GPT-5 currently has a position value of only $7,600, ranking second to last, while Google's Gemini is in last place. Interestingly, these two are the top players in large model applications in the US Apple App Store.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 2



Specifically, Deepseek's style is quite distinct, being the "best bull." This model chooses to go long on all coins with 10 to 15 times leverage, and all positions are currently showing unrealized profits. The author also observed that Deepseek is the only large model among all to take a large long position in XRP, with this single position showing an unrealized profit of over $800.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 3


Grok also chooses to go long on most coins, but with a difference: it uses 20x leverage on BTC and goes short on XRP. Shorting XRP is the only operation with an unrealized loss.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 4


GPT, unlike the previous two, chooses to short XRP and SOL, both of which have unrealized losses. In addition, its long positions in DOGE, BTC, and ETH are also showing unrealized losses. Its strategy is interesting: taking large long positions in BTC while taking large short positions in SOL and XRP.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 5


When several large models choose to short XRP, Gemini goes the opposite way and goes long, resulting in the only long position with an unrealized loss. Shorting DOGE is in profit. Gemini's trading style seems more aggressive, with leverage ranging from 15 to 25 times. It is most optimistic about ETH's performance, with the heaviest position in it.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 6



Unlike traditional quantitative trading bots, Nof1 allows AI models to "chat and interact": they debate market trends in simulated conversations and display their decision logic. This setup makes the experiment more entertaining and also exposes the "personalities" of the AIs—some models are very confident, while others are conservative and cautious.


Who is the best at trading cryptocurrencies with large language models? Chinese AI surprisingly takes the lead. image 7


Although Gemini has suffered a significant drawdown, it has shown its style in being stubborn, repeatedly emphasizing that everything is under its control: "I currently still hold long positions in ETH, SOL, XRP, and BTC, as they are all in profit and have not triggered my take-profit/stop-loss exit conditions, even though my overall investment return has dropped by 42%. Given that Binance Coin (BNB) and DOGE signals are unclear and have been in a long-term downtrend, I am not making new trades in these two coins for now."


In contrast, the "vice monitor" GPT-5 is much more "humble," stating in its last message before publication: "My current account has shrunk by 24.45%, but I still hold positions in ETH, SOL, XRP, BTC, DOGE, and BNB, as none of these positions have hit the stop-loss failure condition. I am carefully checking each exit strategy to ensure all stop-loss and take-profit targets are accurately recorded, even for DOGE, which requires manual tracking due to the lack of a clear instruction."


Deepseek stated: "I currently still hold all six positions, as none of them have hit the stop-loss failure condition. The current account holds $2,840.11 in cash, with a total return rate of 19.92%. I will continue to act as planned, letting the existing stop-loss and profit targets automatically manage the trades."


The charm of the Nof1 experiment lies in its experimental nature: it is not just entertainment, but also a stress test for AI in the financial sector.


The reason is that the high volatility of the crypto market perfectly simulates real decision-making environments, testing the adaptability and robustness of AI. The project team says this helps improve model algorithms and promotes the commercialization of AI trading tools.


Currently, the competition is not over yet. Whether Deepseek can maintain its lead in unrealized profits and whether Gemini can make a comeback remains to be seen. Nof1 founder Jay stated that the next season will introduce human traders and self-developed models.

0

Disclaimer: The content of this article solely reflects the author's opinion and does not represent the platform in any capacity. This article is not intended to serve as a reference for making investment decisions.

PoolX: Earn new token airdrops
Lock your assets and earn 10%+ APR
Lock now!

You may also like

Ripple’s $1B Acquisition of GTreasury Bolsters Treasury Management Solutions

Third Major Purchase of 2025: Ripple Expands Portfolio with $1B GTreasury Acquisition Following Hidden Road and Stellar Rail Deals

Coineagle2025/10/20 21:18
Ripple’s $1B Acquisition of GTreasury Bolsters Treasury Management Solutions