Anthropic's Claude Fable 5.1 and related Opus/Mythos variants currently lead Humanity's Last Exam (HLE) leaderboards at 59–65% accuracy on the 2,500-question expert-vetted benchmark covering graduate-level math, sciences, and humanities. Recent releases from Anthropic, Meta's Muse Spark, and OpenAI's GPT-5.4/GPT-6 lines have driven rapid gains through enhanced reasoning scaffolds, tool integration, and larger-scale training, outpacing earlier 2025 text-only frontiers below 20%. The benchmark's design keeps saturation low, with top scores clustered tightly and substantial headroom remaining. Key swing factors for year-end 2026 include further Anthropic or OpenAI model drops, advances in agentic workflows, and any verified jumps from Chinese labs like GLM or DeepSeek. Traders weigh these against typical release cadences and the risk of evaluation changes via HLE-Rolling updates.
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$53,587 交易量
60%+
66%
65%以上
32%
70% 以上
17%
75%以上
12%
80%以上
7%
90%+
4%
$53,587 交易量
60%+
66%
65%以上
32%
70% 以上
17%
75%以上
12%
80%以上
7%
90%+
4%
For resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.
The resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to "No".
市場開放時間: Jul 23, 2026, 6:41 PM ET
For resolution, “accuracy” refers solely to the value labeled “HLE Accuracy” or a clear equivalent metric if the data’s presentation or terminology is restructured, regardless of the model’s Calibration Error or any other displayed metric.
The resolution source will be the official Humanity’s Last Exam leaderboard at https://agi.safe.ai/. If the resolution source becomes unavailable during the listed timeframe, this market will remain open to allow the relevant data to become available again. If the source remains unavailable after the end of the listed timeframe or is otherwise confirmed to be permanently unavailable, official Humanity’s Last Exam results published elsewhere may be used. If no official alternative source is available, this market will resolve to "No".
Anthropic's Claude Fable 5.1 and related Opus/Mythos variants currently lead Humanity's Last Exam (HLE) leaderboards at 59–65% accuracy on the 2,500-question expert-vetted benchmark covering graduate-level math, sciences, and humanities. Recent releases from Anthropic, Meta's Muse Spark, and OpenAI's GPT-5.4/GPT-6 lines have driven rapid gains through enhanced reasoning scaffolds, tool integration, and larger-scale training, outpacing earlier 2025 text-only frontiers below 20%. The benchmark's design keeps saturation low, with top scores clustered tightly and substantial headroom remaining. Key swing factors for year-end 2026 include further Anthropic or OpenAI model drops, advances in agentic workflows, and any verified jumps from Chinese labs like GLM or DeepSeek. Traders weigh these against typical release cadences and the risk of evaluation changes via HLE-Rolling updates.
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