Gemini 3.1 Pro Preview's leading 46.4% score on the Humanity's Last Exam leaderboard—edging OpenAI's GPT-5.4 Pro at 44.3%—anchors trader consensus at 59% implied probability for Gemini surpassing 50% by June 30, per Polymarket odds. This frontier benchmark of 2,500 expert-level questions across mathematics, sciences, and humanities highlights Google's advances in reasoning and hallucination reduction via agentic enhancements like Deep Research, as shown in February 2026 evaluations. April's Gemini Enterprise Agent Platform rollout signals ongoing iteration, but leaderboard updates lag new releases. Traders watch Google I/O on May 19-20 for potential Gemini 4 previews or capability demos that could push scores higher amid fierce AI lab competition, though evaluation delays remain a risk.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated$312,073 Vol.
50%+
59%
55%+
28%
60%+
6%
$312,073 Vol.
50%+
59%
55%+
28%
60%+
6%
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...Gemini 3.1 Pro Preview's leading 46.4% score on the Humanity's Last Exam leaderboard—edging OpenAI's GPT-5.4 Pro at 44.3%—anchors trader consensus at 59% implied probability for Gemini surpassing 50% by June 30, per Polymarket odds. This frontier benchmark of 2,500 expert-level questions across mathematics, sciences, and humanities highlights Google's advances in reasoning and hallucination reduction via agentic enhancements like Deep Research, as shown in February 2026 evaluations. April's Gemini Enterprise Agent Platform rollout signals ongoing iteration, but leaderboard updates lag new releases. Traders watch Google I/O on May 19-20 for potential Gemini 4 previews or capability demos that could push scores higher amid fierce AI lab competition, though evaluation delays remain a risk.
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