- Will the highest score achieved by an OpenAI model on Humanity’s Last Exam in 2026 be 50% or higher?95¢
- Will the highest score achieved by an OpenAI model on Humanity’s Last Exam in 2026 be 55% or higher?85¢
- Will the highest score achieved by an OpenAI model on Humanity’s Last Exam in 2026 be 60% or higher?54¢
- Will the highest score achieved by an OpenAI model on Humanity’s Last Exam in 2026 be 65% or higher?34¢
- Will the highest score achieved by an OpenAI model on Humanity’s Last Exam in 2026 be 70% or higher?11¢
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 and Opus 5 currently lead the Humanity’s Last Exam leaderboard at 55.5% and 54.9% respectively as of early September 2026, while OpenAI’s strongest entry, GPT-5.6 Sol, sits at 49.5%. The 2,500-question benchmark, created by the Center for AI Safety and Scale AI, tests graduate-level expertise across math, physics, biology, and other domains with questions designed to resist quick saturation. OpenAI’s GPT-5 series has shown steady gains through improved reasoning chains and post-training, yet trails on the text-only, no-tools version that defines the main leaderboard. Further OpenAI model releases, agentic scaffolding, or verified tool-use evaluations before December 31 could shift relative positioning, as the frontier has climbed from single-digit scores in early 2025 to the mid-50s in roughly 18 months.