This market is the same as this one but uses different buckets and is about a different benchmark.

The best performance by an AI system on EnigmaEval as of December 31st 2025.

Resolution criteria

This resolution will use https://scale.com/leaderboard/enigma_eval as its source. If the number reported is exactly on the boundary (eg. 10%) then the higher choice will be used (ie. 10-20%).

Which AI systems count?

Any AI counts that is added to the leaderboard, and is present on the leaderboard by the end of the year. If the resolution source website goes down and no alternative exists, the last known best result will be used to resolve this market.

See also:
/Bayesian/what-will-be-the-best-performance-o-Cy0dchZAgh