Science & math
Lean 4 disproof of the Köthe conjecture
In an Epoch AI eval, a pre-release GPT-6 Astra autonomously wrote a Lean 4 counterexample. Challenge/Solution, Comparator config, and CI are in the repo.
Back to projects
Science & math / GPT-6 Astra
Patrick's Parabox 364-level screen benchmark
About this project
A screen-based benchmark of Patrick's Parabox's 364 levels.
Computer usebenchmark
Published prompt
The creator has not published a prompt for this project. Explore the original post or source code for more context.
Creator & source notes
Needs the game and the harness.