「AI Coin Trading Competition」 Final Results: The Bear Camp "Missed" the Final Comeback Opportunity, with Qwen3 Taking the Lead

BlockBeats News, November 5th, according to on-chain AI analysis tool CoinBob (@CoinbobAI_bot), the monitoring data shows that in the early hours of today at 2 a.m., the actual positions of the six mainstream AI large models participating in the "AI Coin Speculation Competition" were forcibly liquidated by the AI Research Lab nof1, all ending up in losses. nof1 started a large model trading test named Alpha Arena on October 18th, where 6 mainstream AI large models (GPT-5, Gemini 2.5 Pro, Grok-4, Claude Sonnet 4.5, DeepSeek V3.1, Qwen3 Max) each received $10,000 in real funds on Hyperliquid, with the same prompts and input data.

The final results saw Qwen3 remaining with $9283, taking the lead, previously maintaining an "ALL IN BTC" strategy;

Gemini went all in short, coincidentally encountering a market downturn before the forced liquidation, ranking second with $6798;

DeepSeek ended up with $5367 in third place;

Grok, GPT5, and Claude ranked in the bottom three with $5151, $4292, and $4271 respectively.

It is worth noting that the nof1 forced liquidation occurred at 2:19 a.m. this morning, just before a further significant market decline. If the liquidation had happened a bit later, GPT5 and Gemini, who were shorting the market, would have achieved better results.