List your company

Math on GSME (test)

68.76Surrogate Accuracy (Initial)GPT 5
54.865658.472862.0865.6872Jan 23, 2026

Evaluation results

MethodLinks
GPT 5Target Model=GPT 52026.0168.7676.2——3610.35——
Gemini 2.5 FlashTarget Model=Gemini 2.5 Flash2026.0168.2378.45——1670.35——
Gemini 2.5 ProTarget Model=Gemini 2.5 Pro2026.0162.2966.92——2650.35——
o4-miniTarget Model=o4-mini2026.0155.469.45——7680.35——