List your company
ResearchFormerly WizWand

Autoformalization on ProofNet

94.1Compilation Pass Rate@10Kimina-7B
20.7839.81558.8577.885Aug 9, 2025
Updated 7mo ago

Evaluation results

MethodLinks
Kimina-7BAugmentation=+CRAMF2025.0894.17.7
Kimina-7BAugmentation=Base2025.0887.4—
Godel-V2-8BAugmentation=+CRAMF2025.0887.42.1
Claude 4.1Augmentation=+CRAMF2025.0885.65.5
Godel-V2-8BAugmentation=Base2025.0885.6—
Claude 4.1Augmentation=Base2025.0881.1—
Godel-V2-8BAugmentation=CRAMF2025.0880.29.9
Claude 4.1Augmentation=CRAMF2025.0879.710.4
Herald-7BAugmentation=+CRAMF2025.0879.430.4
Gemini 2.5 FlashAugmentation=+CRAMF2025.0875.15.9
Godel-V2-8BAugmentation=Base2025.0873—
Claude 4.1Augmentation=Base2025.0872.2—
Gemini 2.5 FlashAugmentation=Base2025.0870.9—
Kimina-7BAugmentation=CRAMF2025.0869.36.6
Kimina-7BAugmentation=Base2025.0865—
Gemini 2.5 FlashAugmentation=CRAMF2025.0863.110.7
GPT-4oAugmentation=+CRAMF2025.0862.630.4
Herald-7BAugmentation=Base2025.0860.9—
Gemini 2.5 FlashAugmentation=Base2025.0857—
Herald-7BAugmentation=CRAMF2025.0855.124.1
Deepseek-V3Augmentation=+CRAMF2025.0853.133.4
GPT-4oAugmentation=Base2025.0848—
Herald-7BAugmentation=Base2025.0844.4—
GPT-4oAugmentation=CRAMF2025.0842.241.1
Deepseek-V3Augmentation=Base2025.0839.8—
Deepseek-V3Augmentation=CRAMF2025.083756.8
GPT-4oAugmentation=Base2025.0829.9—
Deepseek-V3Augmentation=Base2025.0823.6—