Highest Claude score on Humanity’s Last Exam in 2026?
- 55%+
- 86%
- +12.0
- 60%+
- 54%
- +1.5
- 65%+
- 15%
- —
- 70%+
- 8%
- −1.5
- 75%+
- 5%
- −1.3
Traders currently price "55%+" at 86% for "Highest Claude score on Humanity’s Last Exam in 2026?". Over the past 24 hours that probability moved up 12.0 points. Full distribution: 55%+ 86%, 60%+ 54%, 65%+ 15%, 70%+ 8%. The market has traded $58K in total volume and resolves Dec 31, 2026. Anthropic’s July 2026 release of Claude Opus 5 and related variants, including Claude Fable 5, quickly established the current frontier on Humanity’s Last Exam, a demanding 2,500-question benchmark spanning expert-level topics in math, sciences, and humanities. These models posted leading scores of roughly 55–65% depending on evaluation settings and inference optimizations such as adaptive reasoning, outpacing GPT-5.6 variants near 50% and other competitors. Trader sentiment reflects Anthropic’s rapid iteration cadence, strong post-training techniques, and sustained edge in long-horizon reasoning amid a benchmark designed to resist quick saturation. Key remaining catalysts through year-end include further Claude fine-tunes, larger training runs, or new Opus releases that could push the 2026 high-water mark higher while rivals close gaps.
Odds sourced live from Polymarket, refreshed through the day. Prediction-market participation carries risk and may be restricted in your jurisdiction. Informational only — not investment advice.