List your company

Outcome Reasoning on Arithmetic

87.8M' F1 MeanGPT-5
57.95265.70173.4581.199May 17, 2025

Evaluation results

MethodLinks
GPT-5Model=GPT-52025.0587.882.7
GPT-o4Model=GPT-o42025.0585.880.9
Llama4-MModel=Llama4-M2025.0576.370.6
DeepSeekModel=DeepSeek2025.057467.5
Gemini2.5Model=Gemini2.52025.0572.165.4
Qwen3Model=Qwen32025.0569.863.2
Llama4-SModel=Llama4-S2025.0559.152.4