Anthropic's Claude Opus 4.7 release in mid-April 2026 propelled its score to 36.2% on the official Scale AI Humanity's Last Exam leaderboard—a rigorous benchmark of 2,500 expert-vetted questions testing frontier AI reasoning across math, science, and humanities—surpassing prior Claude Opus 4.6 at 34.4% and solidifying trader consensus for thresholds below 40%. This reflects steady scaling gains amid intense competition, with Google's Gemini 3.1 Pro Preview leading at 46.4% and OpenAI's GPT-5.4 close behind. While internal previews like Claude Mythos hint at 60%+ potential with tools, public leaderboard listings lag due to evaluation protocols. Traders eye June model drops or leaderboard updates as catalysts, though benchmark saturation on ultra-hard questions tempers odds for 45%+ breakthroughs by June 30.
Экспериментальная сводка, созданная ИИ на основе данных Polymarket. Это не является торговой рекомендацией и не влияет на то, как разрешается этот рынок. · ОбновленоОценка Клода на последнем экзамене человечества к 30 июня?
Оценка Клода на последнем экзамене человечества к 30 июня?
$283,101 Объем
45%+
19%
50%+
9%
55%+
4%
$283,101 Объем
45%+
19%
50%+
9%
55%+
4%
The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Открытие рынка: Jan 30, 2026, 12:00 AM ET
Resolver
0x65070BE91...The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Resolver
0x65070BE91...Anthropic's Claude Opus 4.7 release in mid-April 2026 propelled its score to 36.2% on the official Scale AI Humanity's Last Exam leaderboard—a rigorous benchmark of 2,500 expert-vetted questions testing frontier AI reasoning across math, science, and humanities—surpassing prior Claude Opus 4.6 at 34.4% and solidifying trader consensus for thresholds below 40%. This reflects steady scaling gains amid intense competition, with Google's Gemini 3.1 Pro Preview leading at 46.4% and OpenAI's GPT-5.4 close behind. While internal previews like Claude Mythos hint at 60%+ potential with tools, public leaderboard listings lag due to evaluation protocols. Traders eye June model drops or leaderboard updates as catalysts, though benchmark saturation on ultra-hard questions tempers odds for 45%+ breakthroughs by June 30.
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