🏆 MiniCTX Leaderboard 🏆

MiniCTX evaluates AI theorem provers using real examples from research-level formal math projects.

huggingface

Paper

ICLR '25

In-file context Full repository context Premise selection

Best-of-1:

⚡MiniCTX⚡

📝 Notes

Evaluation was conducted using the code in this repository.
Credit to EvalPlus for the leaderboard site layout.