Anthropic’s latest Claude Fable 5 variant currently leads Humanity’s Last Exam leaderboards at 53.3–64.5% depending on configuration (adaptive reasoning, max effort, tools), outpacing Gemini 3.1 Pro Preview and GPT-5.4 models. This positioning stems from recent releases emphasizing extended thinking modes and tool integration on the 2,500-question benchmark, which tests frontier reasoning across disciplines. With only two weeks until the June 30 resolution, traders are focused on whether incremental optimizations or a new Claude iteration can push scores into higher buckets before the deadline. Competitive pressure from OpenAI and Google releases adds uncertainty, though Anthropic holds the edge in published results.
基於Polymarket數據的AI實驗性摘要。這不是交易建議,也不影響該市場的結算方式。 · 更新於$363,753 交易量
45%+
46%
50%以上
17%
55%+
10%
$363,753 交易量
45%+
46%
50%以上
17%
55%+
10%
The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
市場開放時間: 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 latest Claude Fable 5 variant currently leads Humanity’s Last Exam leaderboards at 53.3–64.5% depending on configuration (adaptive reasoning, max effort, tools), outpacing Gemini 3.1 Pro Preview and GPT-5.4 models. This positioning stems from recent releases emphasizing extended thinking modes and tool integration on the 2,500-question benchmark, which tests frontier reasoning across disciplines. With only two weeks until the June 30 resolution, traders are focused on whether incremental optimizations or a new Claude iteration can push scores into higher buckets before the deadline. Competitive pressure from OpenAI and Google releases adds uncertainty, though Anthropic holds the edge in published results.
基於Polymarket數據的AI實驗性摘要。這不是交易建議,也不影響該市場的結算方式。 · 更新於
警惕外部連結哦。
警惕外部連結哦。
Frequently Asked Questions