Google's Gemini 3.1 Pro Preview currently tops the Humanity's Last Exam leaderboard at 46.44% accuracy in high thinking mode—edging out OpenAI's GPT-5.4 Pro (44.32%) and Meta's Muse Spark (40.56%)—reflecting recent post-training gains in frontier reasoning across 2,500 expert-vetted, multi-modal questions spanning math, sciences, and humanities. This leadership, confirmed on Scale Labs evaluations finalized April 2025, underscores Gemini's competitive positioning amid rapid benchmark progress noted in Stanford's 2026 AI Index, where top scores have surged from under 10% to over 45%. Traders monitor Google I/O (May 19-20) for potential Gemini 3.2 releases or agentic upgrades like Deep Research Max (reported 54.6% variant), which could push scores higher before the June 30 deadline, though calibration errors highlight persistent overconfidence risks.
Экспериментальная сводка, созданная ИИ на основе данных Polymarket. Это не является торговой рекомендацией и не влияет на то, как разрешается этот рынок. · ОбновленоОценка Google Gemini на последнем экзамене человечества к 30 июня?
Оценка Google Gemini на последнем экзамене человечества к 30 июня?
$312,073 Объем
50%+
55%
55%+
27%
60%+
6%
$312,073 Объем
50%+
55%
55%+
27%
60%+
6%
The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Открытие рынка: 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 currently tops the Humanity's Last Exam leaderboard at 46.44% accuracy in high thinking mode—edging out OpenAI's GPT-5.4 Pro (44.32%) and Meta's Muse Spark (40.56%)—reflecting recent post-training gains in frontier reasoning across 2,500 expert-vetted, multi-modal questions spanning math, sciences, and humanities. This leadership, confirmed on Scale Labs evaluations finalized April 2025, underscores Gemini's competitive positioning amid rapid benchmark progress noted in Stanford's 2026 AI Index, where top scores have surged from under 10% to over 45%. Traders monitor Google I/O (May 19-20) for potential Gemini 3.2 releases or agentic upgrades like Deep Research Max (reported 54.6% variant), which could push scores higher before the June 30 deadline, though calibration errors highlight persistent overconfidence risks.
Экспериментальная сводка, созданная ИИ на основе данных Polymarket. Это не является торговой рекомендацией и не влияет на то, как разрешается этот рынок. · Обновлено
Не доверяй внешним ссылкам.
Не доверяй внешним ссылкам.
Часто задаваемые вопросы