Anthropic's Claude Opus 4.6 and Sonnet 4.6, released in February 2026, have vaulted to top spots on Humanity's Last Exam (HLE), a rigorous multi-modal benchmark with 2,500 expert-vetted questions probing AI reasoning frontiers across math, science, and humanities. These models score around 33% without tools and up to 49% with agentic enhancements, outpacing rivals like OpenAI's GPT-5.2 and Google's Gemini 3 in key evals, driven by advances in long-context reasoning and tool integration. No major updates since, but traders monitor for Claude 5 previews or mid-year releases per Anthropic's cadence, alongside leaderboard refreshes from Scale AI or Artificial Analysis—timelines could slip amid scaling challenges and safety reviews.
Экспериментальная сводка, созданная ИИ на основе данных Polymarket · ОбновленоАнтропный балл Клода на последнем экзамене человечества к 30 июня?
Антропный балл Клода на последнем экзамене человечества к 30 июня?
$187,336 Объем
35%+
94%
45%+
49%
$187,336 Объем
35%+
94%
45%+
49%
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 4.6 and Sonnet 4.6, released in February 2026, have vaulted to top spots on Humanity's Last Exam (HLE), a rigorous multi-modal benchmark with 2,500 expert-vetted questions probing AI reasoning frontiers across math, science, and humanities. These models score around 33% without tools and up to 49% with agentic enhancements, outpacing rivals like OpenAI's GPT-5.2 and Google's Gemini 3 in key evals, driven by advances in long-context reasoning and tool integration. No major updates since, but traders monitor for Claude 5 previews or mid-year releases per Anthropic's cadence, alongside leaderboard refreshes from Scale AI or Artificial Analysis—timelines could slip amid scaling challenges and safety reviews.
Экспериментальная сводка, созданная ИИ на основе данных Polymarket · Обновлено
Не доверяй внешним ссылкам.
Не доверяй внешним ссылкам.
Часто задаваемые вопросы