List your company

Outcome Reasoning on CVQA Count

0.792F1 Mean (M')GPT-5
0.5060.580250.65450.72875May 17, 2025

Evaluation results

MethodLinks
GPT-5Model=GPT-52025.050.7920.72
GPT-o4Model=GPT-o42025.050.7620.693
Llama4-MModel=Llama4-M2025.050.6750.61
DeepSeekModel=DeepSeek2025.050.6350.569
Qwen3Model=Qwen32025.050.6220.557
Gemini2.5Model=Gemini2.52025.050.6010.538
Llama4-SModel=Llama4-S2025.050.5170.452