资讯

CDOTrends12 小时
New AI Models Push the Envelope Again
As reported by TechCrunch, Grok 4 scored 25.4% on Humanity’s Last Exam without “tools,” outperforming Google’s Gemini 2.5 Pro ...