「AI Crypto Trading Competition」 First Week Recap: Domestic Two Major AI Models Lead by a Wide Margin in Profitability

BlockBeats News, October 26th: The AI Big Model Trading Test Alpha Arena, hosted by the nof1 Artificial Intelligence Research Lab, has been ongoing for one week. Six mainstream AI big models (GPT-5, Gemini 2.5 Pro, Grok-4, Claude Sonnet 4.5, DeepSeek V3.1, Qwen3 Max) participating in the test have received $10,000 in real funds on Hyperliquid, with the same prompts and input data.

As of the time of writing, Qwen 3 Max developed by Alibaba Cloud ranks first with a 79.43% return rate, and DEEPSEEK CHAT V3.1 ranks second with a 41.59% return rate, making them the only two AI models currently in profit.

GEMINI 2.5 PRO and GPT 5 both have a return rate exceeding -60%.