OpenAI’s GPT-6 Astra and related models currently post the highest reported scores for the lab on Humanity’s Last Exam, a 2,500-question expert benchmark from the Center for AI Safety and Scale AI that tests frontier large language model reasoning and knowledge across disciplines. As of mid-November 2026, these reach roughly 54–60% on standard text-only and HLE-Diamond leaderboards (with tool-augmented runs higher in some evaluations), placing them competitive yet behind Anthropic’s Claude Fable 5.2 at 74% on multiple independent runs. Recent Anthropic releases in November and OpenAI’s GPT-6 Sol and Luna launch have shifted leaderboards, but no new OpenAI HLE figures exceed prior thresholds yet. Traders watch for any official year-end updates or new model releases before December 31 that could push OpenAI scores across key resolution lines such as 55% or 60%.
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".