external benchmark
ProofBench leaderboard
Vals ProofBench. Every model the catalog carries a published ProofBench value for, ranked by that value.
Categoryexternal
MeasureAccuracy score
TasksAutomated theorem proving
DifficultyFormal proof reasoning
Published byProofBench
Vals AI automated theorem-proving benchmark.
No model in the catalog has a published ProofBench score.
The benchmark is defined by ProofBench, but the catalog carries no value for it yet. An absent value is shown as absent here rather than as a zero.