AI model scores ≥ 90% on FrontierMath Benchmark before 2027?

Traders currently price "Yes" at 89% for "AI model scores ≥ 90% on FrontierMath Benchmark before 2027?". Over the past 24 hours that probability moved down 0.5 points. The market has traded $117K in total volume and resolves Dec 31, 2026. Recent releases of frontier large language models, particularly OpenAI's GPT-5.6 Sol reaching 89% on the legacy FrontierMath leaderboard and strong results near 83% on the updated v2 tier released in June 2026, have driven the 89% market-implied odds for crossing 90% before 2027. Ongoing advances in mathematical reasoning, tool use, and post-training techniques across labs like OpenAI, Anthropic, and others continue to close the gap on this benchmark of unpublished, expert-level problems. With several new model iterations expected in the coming months and historical patterns of rapid benchmark gains, traders see limited barriers remaining despite the benchmark's difficulty.

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.