AI model scores ≥ 90% on FrontierMath Benchmark before 2027?
BUY YES
- Cost
- 88.0¢
- Implied chance
- 88.0%
- Return if right
- 13.6%
- Spread cost
- 5%
If you put in $100
You get 113.6 shares at 88.0¢ each.
YES wins: $113.64 back, a profit of $13.64. YES loses: $0.00, and the whole $100 is gone.
The gap between buyers and sellers takes $0.65 of that profit. Without it you would make $14.29.
Assumes the whole order fills at this price. A large order eats into the book and averages worse.
Buy YES on Polymarket
BUY NO
- Cost
- 13.0¢
- Implied chance
- 13.0%
- Return if right
- 669.2%
- Spread cost
- 4%
If you put in $100
You get 769.2 shares at 13.0¢ each.
NO wins: $769.23 back, a profit of $669.23. NO loses: $0.00, and the whole $100 is gone.
The gap between buyers and sellers takes $30.77 of that profit. Without it you would make $700.00.
Assumes the whole order fills at this price. A large order eats into the book and averages worse.
Buy NO on Polymarket
Latest news
headlines link to their publishersNo recent headlines matched this market. We only show an article when the names in the question appear in its headline, so a very local or very new market often has nothing. Showing nothing is deliberate: an unrelated story here would read like evidence about the price.
Headlines are matched on the names in “AI model scores ≥ 90% on FrontierMath Benchmark before 2027?”. Nobody checks them by hand, so one may not be relevant. They also say nothing about whether the price has already moved on this news.
How this market resolves
This market will resolve to "Yes" if a state-of-the-art (SOTA) AI model achieves a score of 90% or greater on the FrontierMath Exam by December 31, 2026, 11:59 PM ET. Otherwise, the market will resolve to "No". The primary resolution source will be information from EpochAI however a consensus of credible reporting may also be used.