List your company

Formal Theorem Proving on Ineq-Comp (test)

66.7Ineq-Comp (Seed)DeepSeek-Prover-V2-7B
39.45246.52653.660.674May 21, 2026

Evaluation results

MethodLinks
DeepSeek-Prover-V2-7BBudget=642026.0566.737.355.9
DeepSeek-Prover-V2-7BBudget=322026.0564.834.853.7
DeepSeek-Prover-V2-7B + EnsembleBudget=642026.0564.745.570.3
DeepSeek-Prover-V2-7B + EnsembleBudget=322026.0563.440.864.4
EvolProverBudget=322026.0552.23465.2
Goedel-Prover-DPO + EnsembleBudget=642026.0546.61736.5
Goedel-Prover-DPOBudget=642026.054415.334.8
Goedel-Prover-DPO + EnsembleBudget=322026.0542.314.534.3
Goedel-Prover-DPOBudget=322026.0540.512.330.4