Highest Claude score on Humanity’s Last Exam in 2026?
- Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 55% or higher?84¢
- Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 60% or higher?51¢
- Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 65% or higher?15¢
- Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 70% or higher?8¢
- Will the highest score achieved by an Anthropic Claude model on Humanity’s Last Exam in 2026 be 75% or higher?5¢
This market will resolve to "Yes" if any Anthropic Claude 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 latest Claude releases, including Opus 5 and Fable 5 variants, currently lead Humanity’s Last Exam leaderboards with scores of 54.9–64.7% as of mid-August 2026, reflecting strong trader consensus that the company’s emphasis on advanced reasoning and adaptive techniques gives it an edge on this expert-level benchmark of 2,500 questions. Rapid progress since the benchmark’s Nature publication in January 2026 has lifted frontier models from single-digit accuracy to the mid-50s or higher, driven by architectural improvements targeting complex knowledge and multi-step inference. Competitive pressure from OpenAI’s GPT series and Google’s Gemini remains intense, with clusters of top scores within a few points suggesting further gains depend on scaling, post-training methods, and any new model launches before year-end. Traders are watching Anthropic’s next releases and benchmark updates closely, as small capability shifts could determine whether a Claude model retains the highest mark through December 2026.