BlockBeats News, October 20th – The AI research lab nof1, focused on the financial market, started a large-scale model trading experiment named Alpha Arena on the 18th. The test utilizes 6 mainstream AI large models (GPT-5, Gemini 2.5 Pro, Grok-4, Claude Sonnet 4.5, DeepSeek V3.1, Qwen3 Max), with each model receiving $10,000 in real funds on Hyperliquid and having the same prompts and input data.
As of the time of writing, DeepSeek, Grok, and Claude are ranked in the top three with returns of 40.14%, 35.49%, and 24.54% respectively, while Gemini 2.5 Pro is at a floating loss of 30.46%.
