Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MiniF2F

Benchmarks

Task NameDataset NameSOTA ResultTrend
Formal Theorem ProvingMiniF2F (test)
Pass@199.6
128
AutoformalizationMiniF2F
Compilation Pass Rate@10100
28
Formal-to-formal theorem provingminiF2F (test)
Proven Theorems (%)26.5
6
Formal Math ProvingMiniF2F Lean4 (test)
Pass@16 (Overall)69.7
2
Showing 4 of 4 rows