Recent releases of OpenAI’s GPT-5.5 series in April 2026 have driven steady gains on Humanity’s Last Exam, a 2,500-question benchmark of expert-level academic problems across math, science, and humanities. Current leading GPT-5.5 variants reach roughly 57 percent accuracy, placing them close behind Anthropic’s Claude Mythos Preview at 65 percent while ahead of most Google Gemini entries. Traders are watching for any additional fine-tuning, larger context windows, or agentic enhancements OpenAI could deploy before the June 30 resolution cutoff. Competitive pressure from Anthropic’s rapid iteration and Google’s frontier models remains the main swing factor, as even modest capability jumps on this hard benchmark can shift market-implied odds quickly.
Eksperymentalne podsumowanie AI odwołujące się do danych Polymarket. To nie jest porada handlowa i nie ma wpływu na rozstrzyganie tego rynku. · ZaktualizowanoOpenAI GPT score on Humanity’s Last Exam by June 30?
$23,154 Wol.
50%+
30%
$23,154 Wol.
50%+
30%
The resolution source will be the official Humanity’s Last Exam leaderboard https://scale.com/leaderboard/humanitys_last_exam.
Rynek otwarty: 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...Recent releases of OpenAI’s GPT-5.5 series in April 2026 have driven steady gains on Humanity’s Last Exam, a 2,500-question benchmark of expert-level academic problems across math, science, and humanities. Current leading GPT-5.5 variants reach roughly 57 percent accuracy, placing them close behind Anthropic’s Claude Mythos Preview at 65 percent while ahead of most Google Gemini entries. Traders are watching for any additional fine-tuning, larger context windows, or agentic enhancements OpenAI could deploy before the June 30 resolution cutoff. Competitive pressure from Anthropic’s rapid iteration and Google’s frontier models remains the main swing factor, as even modest capability jumps on this hard benchmark can shift market-implied odds quickly.
Eksperymentalne podsumowanie AI odwołujące się do danych Polymarket. To nie jest porada handlowa i nie ma wpływu na rozstrzyganie tego rynku. · Zaktualizowano
Uważaj na linki zewnętrzne.
Uważaj na linki zewnętrzne.
Często zadawane pytania