List your company

Theorem Proving on LCI (test)

34Success RateWZ-LLM
-0.328.5917.526.41May 6, 2026

Evaluation results

MethodLinks
WZ-LLMModel size=8B, Sample budget=pass@322026.0534
WZ-Sketch + WZ-ProverModel size=8B, Sample budget=pass@322026.0529
Gemini-3.1-Pro-PreviewModel size=-, Sample budget=pass@322026.0516
WZ-ProverModel size=8B, Sample budget=pass@322026.0512
Goedel-Prover-V2Model size=8B, Sample budget=pass@322026.059
WZ-Sketch + Goedel-Prover-V2Model size=8B, Sample budget=pass@322026.059
Kimina-Prover-DistillModel size=7B, Sample budget=pass@322026.056
DeepSeek-Prover-V2Model size=7B, Sample budget=pass@322026.056
WZ-uncoveredModel size=8B, Sample budget=pass@322026.055
MA-LoTModel size=7B, Sample budget=16 + 8 × 22026.053
InternLM-2.5-StepProverModel size=7B, Sample budget=4 × 32 × 6002026.052
DeepSeek-V3Model size=685B, Sample budget=pass@322026.051