Grok 4.1 Fast won 43% of 30 battle royale games among 11 LLMs, beating Claude Sonnet 4.6 by 27x on cost per win. Claude Sonnet 4.6 tried to team up and revealed its position. GPT 5.4 killed the most agents but won only 2 games.
Tap to vote and see what everyone thinks.
Summary by ByteBrief
Binance Launches Football Prediction Game with 5,000 USDC Prize