List your company

Formal Theorem Proving on miniF2F rw (val)

81.1Pass@8Goedel-Prover-V2-8B
50.4258.38566.3574.315May 21, 2026

Evaluation results

MethodLinks
Goedel-Prover-V2-8BVariant=seed2026.0581.183.2—68.882.5
Goedel-Prover-V2-8BVariant=controlled2026.0580.682.8——81.9
Goedel-Prover-V2-8BVariant=test-time2026.0580.482.8——81.9
DeepSeek-Prover-V2-7BVariant=test-time2026.0579.781.181.5——
Goedel-Prover-V2-8BVariant=random2026.0579.582.2—65.281.2
DeepSeek-Prover-V2-7BVariant=seed2026.0578.479.179.5——
DeepSeek-Prover-V2-7BVariant=seed2026.0578.180.7—67.479.7
DeepSeek-Prover-V2-7BVariant=controlled2026.0577.781——79.5
DeepSeek-Prover-V2-7BVariant=test-time2026.0577.581——79.3
DeepSeek-Prover-V2-7BVariant=controlled2026.0577.279.780.8——
DeepSeek-Prover-V2-7BVariant=random2026.0575.979.3—62.777.7
Kimina-Prover-Distill-8BVariant=controlled2026.0574.278.3——76.5
Kimina-Prover-Distill-8BVariant=seed2026.0573.476.6—61.975.4
Kimina-Prover-Distill-8BVariant=test-time2026.057276.3——74.5
DeepSeek-Prover-V2-7BVariant=random2026.0571.174.175.3——
Kimina-Prover-Distill-8BVariant=random2026.0570.975.8—57.673.6
Goedel-Prover-DPOVariant=seed2026.0558.663.164.3——
Goedel-Prover-SFTVariant=seed2026.0557.762.864.8——
Goedel-Prover-DPOVariant=controlled2026.0557.763.465.4——
Goedel-Prover-DPOVariant=test-time2026.0557.763.565.2——
Goedel-Prover-SFTVariant=test-time2026.0556.362.663.6——
Goedel-Prover-SFTVariant=controlled2026.0554.760.863.4——
Goedel-Prover-DPOVariant=random2026.0553.959.761.6——
Goedel-Prover-SFTVariant=random2026.0551.658.260.9——