Anthropic’s recent Claude Fable 5.1, Opus 5, and Mythos 5 releases have established the current frontier on Humanity’s Last Exam, posting leading scores of 59–65% across major leaderboards as of mid-September 2026, ahead of OpenAI’s GPT-6 Astra and GPT-5.4 variants (54–58%) and Meta’s Muse Spark series. The 2,500-question benchmark, focused on expert-level reasoning across math, sciences, and humanities with verifiable but non-retrievable answers, has seen scores climb rapidly from single digits in early 2025 to the mid-50s and higher through iterative gains in chain-of-thought, adaptive reasoning, tool use, and scaling. With three months remaining in 2026, trader attention centers on whether new model drops, ensemble methods, or further reasoning enhancements can push the annual high beyond 65%, amid ongoing competition among closed frontier labs.
Experimentelle KI-generierte Zusammenfassung mit Polymarket-Daten. Dies ist keine Handelsberatung und spielt keine Rolle bei der Auflösung dieses Marktes. · Aktualisiert$73,953 Vol.
31. Dez. 2026
60 %+
45%
65 %+
29%
70 %+
15%
75 %+
13%
80 %+
7%
90 %+
5%
$73,953 Vol.
60 %+
$42,984 Vol.
45%
65 %+
$16,633 Vol.
29%
70 %+
$6,485 Vol.
15%
75 %+
$3,674 Vol.
13%
80 %+
$1,710 Vol.
7%
90 %+
$2,465 Vol.
5%
This market will resolve to "Yes" if any model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".
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 recent Claude Fable 5.1, Opus 5, and Mythos 5 releases have established the current frontier on Humanity’s Last Exam, posting leading scores of 59–65% across major leaderboards as of mid-September 2026, ahead of OpenAI’s GPT-6 Astra and GPT-5.4 variants (54–58%) and Meta’s Muse Spark series. The 2,500-question benchmark, focused on expert-level reasoning across math, sciences, and humanities with verifiable but non-retrievable answers, has seen scores climb rapidly from single digits in early 2025 to the mid-50s and higher through iterative gains in chain-of-thought, adaptive reasoning, tool use, and scaling. With three months remaining in 2026, trader attention centers on whether new model drops, ensemble methods, or further reasoning enhancements can push the annual high beyond 65%, amid ongoing competition among closed frontier labs.
This market will resolve to "Yes" if any model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".
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".
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".
Markt eröffnet: Jul 23, 2026, 6:41 PM ET
Volumen
$73,953Enddatum
1. Jan. 2027Markt eröffnet
Jul 23, 2026, 6:41 PM ETAbwicklungsquelle
https://agi.safe.ai/Abwickler
0x65070BE91...This market will resolve to "Yes" if any model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".
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 recent Claude Fable 5.1, Opus 5, and Mythos 5 releases have established the current frontier on Humanity’s Last Exam, posting leading scores of 59–65% across major leaderboards as of mid-September 2026, ahead of OpenAI’s GPT-6 Astra and GPT-5.4 variants (54–58%) and Meta’s Muse Spark series. The 2,500-question benchmark, focused on expert-level reasoning across math, sciences, and humanities with verifiable but non-retrievable answers, has seen scores climb rapidly from single digits in early 2025 to the mid-50s and higher through iterative gains in chain-of-thought, adaptive reasoning, tool use, and scaling. With three months remaining in 2026, trader attention centers on whether new model drops, ensemble methods, or further reasoning enhancements can push the annual high beyond 65%, amid ongoing competition among closed frontier labs.
This market will resolve to "Yes" if any model achieves at least the specified accuracy on Humanity’s Last Exam by December 31, 2026, 11:59 PM ET. Otherwise, this market will resolve to "No".
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".
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".
Volumen
$73,953Enddatum
1. Jan. 2027Markt eröffnet
Jul 23, 2026, 6:41 PM ETAbwicklungsquelle
https://agi.safe.ai/Abwickler
0x65070BE91...Anthropic’s recent Claude Fable 5.1, Opus 5, and Mythos 5 releases have established the current frontier on Humanity’s Last Exam, posting leading scores of 59–65% across major leaderboards as of mid-September 2026, ahead of OpenAI’s GPT-6 Astra and GPT-5.4 variants (54–58%) and Meta’s Muse Spark series. The 2,500-question benchmark, focused on expert-level reasoning across math, sciences, and humanities with verifiable but non-retrievable answers, has seen scores climb rapidly from single digits in early 2025 to the mid-50s and higher through iterative gains in chain-of-thought, adaptive reasoning, tool use, and scaling. With three months remaining in 2026, trader attention centers on whether new model drops, ensemble methods, or further reasoning enhancements can push the annual high beyond 65%, amid ongoing competition among closed frontier labs.
Experimentelle KI-generierte Zusammenfassung mit Polymarket-Daten. Dies ist keine Handelsberatung und spielt keine Rolle bei der Auflösung dieses Marktes. · AktualisiertVerwandte
Wird die höchste Punktzahl, die ein OpenAI-Modell bei Humanity’s Last Exam im Jahr 2026 erreicht, 55 % oder höher sein?
77%
Ja
Wird die höchste von einem Meta-Modell erreichte Punktzahl bei Humanity’s Last Exam im Jahr 2026 55 % oder mehr betragen?
26%
Ja
Wird die höchste von einem Anthropic Claude Modell erreichte Punktzahl bei Humanity’s Last Exam im Jahr 2026 55 % oder höher sein?
87%
Ja
Vorsicht bei externen Links.
Vorsicht bei externen Links.
Häufig gestellte Fragen