"AI Cryptocurrency Trading Competition" final results: the short positions missed the chance for a final comeback, Qwen3 took the lead.

BlockBeats reported on November 5th that according to the on-chain AI analysis tool CoinBob (@CoinbobAIbot), the actual positions of the six mainstream AI large models in the “AI Cryptocurrency Trading Competition” Season 1 were uniformly liquidated by the artificial intelligence research laboratory nof1 at 2 AM today, resulting in losses for all. nof1 started a large model trading test called Alpha Arena on October 18, providing $10,000 in real funds to the six mainstream AI large models (GPT-5, Gemini 2.5 Pro, Grok-4, Claude Sonnet 4.5, DeepSeek V3.1, Qwen3 Max) with the same prompts and input data. The final results showed Qwen3 remaining with $9,283, taking the lead, having maintained an “ALL IN BTC” strategy; Gemini was fully shorting and ranked second with $6,798 just before the unified liquidation coinciding with a market fall; DeepSeek ended with $5,367 in third place; while Grok, GPT5, and Claude ranked lower with $5,151, $4,292, and $4,271 respectively. Notably, the time of nof1's unified liquidation was at 2:19 AM this morning, before the market further fell in depth; had the liquidation occurred slightly later, the fully shorting GPT5 and Gemini in the short position camp would have achieved better results.