Anthropic’s Claude Fable 5.1 currently leads Humanity’s Last Exam leaderboards at 59–65% depending on the evaluation protocol, outpacing OpenAI’s strongest 2026 entries such as GPT-5.4 Pro near 58.7% and GPT-6 Astra around 54.7%. The benchmark’s 2,500 expert-crafted questions across advanced mathematics, physics, philosophy, and other domains remain far from saturation, rewarding genuine reasoning over retrieval and exposing calibration gaps even in frontier large language models. Rapid gains since the 2025 launch reflect iterative improvements in reasoning chains and agent scaffolding, yet OpenAI trails on recent snapshots amid Anthropic’s stronger showing. Further GPT releases, enhanced tool use, or internal scaling before year-end could shift OpenAI’s relative position on this expert-level test.
Résumé expérimental généré par IA à partir des données Polymarket. Ceci n'est pas un conseil de trading et ne joue aucun rôle dans la résolution de ce marché. · Mis à jour$107,717 Vol.
31 déc. 2026
55 % ou plus
77%
60 %+
33%
65 %+
16%
70 %+
10%
$107,717 Vol.
55 % ou plus
$34,332 Vol.
77%
60 %+
$41,111 Vol.
33%
65 %+
$9,428 Vol.
16%
70 %+
$5,516 Vol.
10%
This market will resolve to "Yes" if any model published by OpenAI 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 Claude Fable 5.1 currently leads Humanity’s Last Exam leaderboards at 59–65% depending on the evaluation protocol, outpacing OpenAI’s strongest 2026 entries such as GPT-5.4 Pro near 58.7% and GPT-6 Astra around 54.7%. The benchmark’s 2,500 expert-crafted questions across advanced mathematics, physics, philosophy, and other domains remain far from saturation, rewarding genuine reasoning over retrieval and exposing calibration gaps even in frontier large language models. Rapid gains since the 2025 launch reflect iterative improvements in reasoning chains and agent scaffolding, yet OpenAI trails on recent snapshots amid Anthropic’s stronger showing. Further GPT releases, enhanced tool use, or internal scaling before year-end could shift OpenAI’s relative position on this expert-level test.
This market will resolve to "Yes" if any model published by OpenAI 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".
Marché ouvert : Jul 23, 2026, 6:53 PM ET
Volume
$107,717Date de fin
1 janv. 2027Marché ouvert
Jul 23, 2026, 6:53 PM ETRésolveur
0x65070BE91...This market will resolve to "Yes" if any model published by OpenAI 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 Claude Fable 5.1 currently leads Humanity’s Last Exam leaderboards at 59–65% depending on the evaluation protocol, outpacing OpenAI’s strongest 2026 entries such as GPT-5.4 Pro near 58.7% and GPT-6 Astra around 54.7%. The benchmark’s 2,500 expert-crafted questions across advanced mathematics, physics, philosophy, and other domains remain far from saturation, rewarding genuine reasoning over retrieval and exposing calibration gaps even in frontier large language models. Rapid gains since the 2025 launch reflect iterative improvements in reasoning chains and agent scaffolding, yet OpenAI trails on recent snapshots amid Anthropic’s stronger showing. Further GPT releases, enhanced tool use, or internal scaling before year-end could shift OpenAI’s relative position on this expert-level test.
This market will resolve to "Yes" if any model published by OpenAI 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".
Volume
$107,717Date de fin
1 janv. 2027Marché ouvert
Jul 23, 2026, 6:53 PM ETRésolveur
0x65070BE91...Anthropic’s Claude Fable 5.1 currently leads Humanity’s Last Exam leaderboards at 59–65% depending on the evaluation protocol, outpacing OpenAI’s strongest 2026 entries such as GPT-5.4 Pro near 58.7% and GPT-6 Astra around 54.7%. The benchmark’s 2,500 expert-crafted questions across advanced mathematics, physics, philosophy, and other domains remain far from saturation, rewarding genuine reasoning over retrieval and exposing calibration gaps even in frontier large language models. Rapid gains since the 2025 launch reflect iterative improvements in reasoning chains and agent scaffolding, yet OpenAI trails on recent snapshots amid Anthropic’s stronger showing. Further GPT releases, enhanced tool use, or internal scaling before year-end could shift OpenAI’s relative position on this expert-level test.
Résumé expérimental généré par IA à partir des données Polymarket. Ceci n'est pas un conseil de trading et ne joue aucun rôle dans la résolution de ce marché. · Mis à jourConnexes
Le score le plus élevé obtenu à l’Ultime Exam de l’Humanité en 2026 sera-t-il de 60% ou plus ?
45%
Oui
Le score le plus élevé obtenu par un modèle Meta à l’ultime examen de l’humanité en 2026 sera-t-il de 55 % ou plus ?
26%
Oui
Le score le plus élevé obtenu par un modèle Anthropic Claude à l'examen final de l'humanité en 2026 sera-t-il de 55 % ou plus ?
86%
Oui
Méfiez-vous des liens externes.
Méfiez-vous des liens externes.
Questions fréquentes