Google's Gemini 3.1 Pro Preview model currently tops the Humanity's Last Exam leaderboard at 44.7%, outpacing OpenAI's GPT-5.4 (41.6%) on this rigorous 2,500-question benchmark testing PhD-level reasoning across mathematics, sciences, and humanities—far below human expert performance near 90% but a leap from sub-30% scores in late 2025. February's Gemini 3 Deep Think release briefly hit 48.4% in early tests, driving sentiment amid fierce competition from Anthropic's Claude series and xAI's Grok. Recent April updates like Gemini 3.1 Flash enhancements signal ongoing iteration, with Google I/O in May poised for major announcements that could push toward 50% thresholds. Traders watch official Scale AI or Artificial Analysis leaderboards for resolution by June 30.
Polymarket verilerine atıfta bulunan deneysel AI tarafından oluşturulmuş özet. Bu bir işlem tavsiyesi değildir ve bu piyasanın nasıl çözümlendiğinde hiçbir rolü yoktur. · Güncellendi$305,941 Hac.
%50+
40%
%55+
20%
%60+
10%
$305,941 Hac.
%50+
40%
%55+
20%
%60+
10%
The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Piyasa Açıldı: Jan 29, 2026, 12:50 PM ET
Resolver
0x65070BE91...The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Resolver
0x65070BE91...Google's Gemini 3.1 Pro Preview model currently tops the Humanity's Last Exam leaderboard at 44.7%, outpacing OpenAI's GPT-5.4 (41.6%) on this rigorous 2,500-question benchmark testing PhD-level reasoning across mathematics, sciences, and humanities—far below human expert performance near 90% but a leap from sub-30% scores in late 2025. February's Gemini 3 Deep Think release briefly hit 48.4% in early tests, driving sentiment amid fierce competition from Anthropic's Claude series and xAI's Grok. Recent April updates like Gemini 3.1 Flash enhancements signal ongoing iteration, with Google I/O in May poised for major announcements that could push toward 50% thresholds. Traders watch official Scale AI or Artificial Analysis leaderboards for resolution by June 30.
Polymarket verilerine atıfta bulunan deneysel AI tarafından oluşturulmuş özet. Bu bir işlem tavsiyesi değildir ve bu piyasanın nasıl çözümlendiğinde hiçbir rolü yoktur. · Güncellendi
Harici bağlantılara dikkat edin.
Harici bağlantılara dikkat edin.
Sıkça Sorulan Sorular