Share your thoughts, 1 month free Claude Pro on usSee more

Autoformalization on AdvancedMath

100Compilation Pass Rate@10

Kimina-7B

Updated 26d ago

Evaluation Results

Method	Links
Kimina-7B 2025.08		100	1.2
Godel-V2-8B 2025.08		100	1.2
Claude 4.1 2025.08		99.4	0.6
Claude 4.1 2025.08		98.8	-
Kimina-7B 2025.08		98.8	-
Godel-V2-8B 2025.08		98.8	-
Herald-7B 2025.08		90.2	12.8
Gemini 2.5 Flash 2025.08		82.1	3.5
Herald-7B 2025.08		80	-
Gemini 2.5 Flash 2025.08		79.3	-
Godel-V2-8B 2025.08		68.2	18
Kimina-7B 2025.08		66.5	8.5
Claude 4.1 2025.08		65.9	15.2
Kimina-7B 2025.08		61.3	-
GPT-4o 2025.08		59.1	21.9
Godel-V2-8B 2025.08		57.8	-
Claude 4.1 2025.08		57.2	-
Herald-7B 2025.08		51.4	30.8
Gemini 2.5 Flash 2025.08		50.9	15.9
GPT-4o 2025.08		48.5	-
Deepseek-V3 2025.08		48.2	52.1
Gemini 2.5 Flash 2025.08		43.9	-
Herald-7B 2025.08		39.3	-
GPT-4o 2025.08		34.7	52.2
Deepseek-V3 2025.08		32.1	62.1
Deepseek-V3 2025.08		31.7	-
GPT-4o 2025.08		22.8	-
Deepseek-V3 2025.08		19.8	-