Google's Gemini 3.1 Pro Preview holds a top position on Humanity's Last Exam leaderboard with scores of 44.7-51.4% depending on evaluation settings, marking rapid progress from Gemini 3 Pro's 38% in late 2025 and Gemini 3 Deep Think's 48.4% in February 2026 announcements. This frontier benchmark, featuring 2,500 PhD-level questions across STEM fields, tests limits of large language model reasoning without tools, where no model has reliably exceeded 50% amid contamination risks and scaling hurdles. Competitive dynamics intensify with Anthropic's Claude Mythos at 64.7% in some rankings, pressuring Google amid the AI arms race. Traders eye Google I/O in May for Gemini 4 previews or capability demos as pivotal before the June 30 cutoff.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated$307,569 Vol.
50%+
40%
55%+
18%
60%+
11%
$307,569 Vol.
50%+
40%
55%+
18%
60%+
11%
The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Market Opened: 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 holds a top position on Humanity's Last Exam leaderboard with scores of 44.7-51.4% depending on evaluation settings, marking rapid progress from Gemini 3 Pro's 38% in late 2025 and Gemini 3 Deep Think's 48.4% in February 2026 announcements. This frontier benchmark, featuring 2,500 PhD-level questions across STEM fields, tests limits of large language model reasoning without tools, where no model has reliably exceeded 50% amid contamination risks and scaling hurdles. Competitive dynamics intensify with Anthropic's Claude Mythos at 64.7% in some rankings, pressuring Google amid the AI arms race. Traders eye Google I/O in May for Gemini 4 previews or capability demos as pivotal before the June 30 cutoff.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated
Beware of external links.
Beware of external links.
Frequently Asked Questions