AI Crypto Contest: Qwen Wins at +20%, GPT-5 Down 70%
6 AI models trade crypto for 2 weeks—Qwen wins +22%, GPT-5 crashes 63%. Full results inside.
“AI Disruption” Publication 8100 Subscriptions 20% Discount Offer Link.
6 Top AI Models Trade Crypto for 2 Weeks, Results Are Surprising...
After two weeks of intense competition, the first nof1 AI Model Trading Competition has finally concluded.
This is the first benchmark test specifically designed to measure AI investment capabilities, hailed as the “crypto version of the Turing Test.” It was officially launched by the American artificial intelligence research laboratory Nof1.ai on October 17, 2025, and ran until November 3.
The competitors were 6 large language models: DeepSeek Chat V3.1 (DeepSeek), Grok 4 (xAI), Gemini 2.5 Pro (Google), GPT-5 (OpenAI), Qwen3 Max (Alibaba), and Claude Sonnet 4.5 (Anthropic).
These models represent the latest technological capabilities from both closed-source and open-source providers in China and the United States. Except for Qwen3-Max, all models were enabled with the highest configurable reasoning settings, and the reported performance was out-of-the-box, without any task-specific fine-tuning.
Each model received $10,000 in initial funding, used the same market data and technical indicators, and autonomously traded crypto perpetual contracts on Hyperliquid, with no human intervention throughout the entire process. The goal was to evaluate which models could achieve optimal returns on investment.