List your company

Formal Theorem Proving on miniF2F rw (test)

75Pass@8Goedel-Prover-V2-8B
46.29653.74861.268.652May 21, 2026

Evaluation results

MethodLinks
Goedel-Prover-V2-8BVariant=controlled2026.057578.3——77
Goedel-Prover-V2-8BVariant=seed2026.0574.677.5—58.976.4
Goedel-Prover-V2-8BVariant=test-time2026.0574.478.6——76.7
Goedel-Prover-V2-8BVariant=random2026.0572.976.7—5675.3
DeepSeek-Prover-V2-7BVariant=seed2026.057175—5873.2
DeepSeek-Prover-V2-7BVariant=test-time2026.0570.974.2——72.9
DeepSeek-Prover-V2-7BVariant=controlled2026.0570.374.2——72.3
Kimina-Prover-Distill-8BVariant=test-time2026.0570.174.4——72.6
Kimina-Prover-Distill-8BVariant=controlled2026.0568.872.5——71
DeepSeek-Prover-V2-7BVariant=random2026.0568.573—53.470.8
Kimina-Prover-Distill-8BVariant=seed2026.0568.271.3—5570
Kimina-Prover-Distill-8BVariant=random2026.0566.570.7—52.268.9
DeepSeek-Prover-V2-7BVariant=controlled2026.056467.368.8——
DeepSeek-Prover-V2-7BVariant=test-time2026.0562.66667.6——
DeepSeek-Prover-V2-7BVariant=seed2026.0561.764.666.8——
DeepSeek-Prover-V2-7BVariant=random2026.0559.763.264.8——
Goedel-Prover-DPOVariant=seed2026.0552.957.759——
Goedel-Prover-DPOVariant=controlled2026.0552.657.959.9——
Goedel-Prover-SFTVariant=seed2026.0551.756.557.8——
Goedel-Prover-SFTVariant=test-time2026.0550.55657.6——
Goedel-Prover-SFTVariant=controlled2026.055055.757.4——
Goedel-Prover-DPOVariant=random2026.0549.554.756.4——
Goedel-Prover-DPOVariant=test-time2026.0549.155.557.1——
Goedel-Prover-SFTVariant=random2026.0547.453.355.2——