List your company
ResearchFormerly WizWand

Outcome Reasoning on CVQA-Bool

81.2M' (F1 Score)GPT-5
53.74460.8726875.128May 17, 2025
Updated 8mo ago

Evaluation results

MethodLinks
GPT-5Model=GPT-52025.0581.274.5
GPT-o4Model=GPT-o42025.0577.170.2
Llama4-MModel=Llama4-M2025.0570.964.3
DeepSeekModel=DeepSeek2025.0566.759.8
Qwen3Model=Qwen32025.0565.458.6
Gemini2.5Model=Gemini2.52025.0563.256.8
Llama4-SModel=Llama4-S2025.0554.848.1