This study investigates the linguistic patterns in human subject descriptions that came out from large language models with vision capabilities. Descriptions generated by two vision-capable LLMs, Qwen2.5-VL-72B-Instruct and Llama3.2-vision:11b, were examinated for a subset of the CelebA dataset. Using word frequency and clustering analyses, we have identified distinct common topics in descriptions of people, including facial features, hair characteristics, clothing, body structure, posture, and contextual environment. Our findings showed differences in how these models organize descriptive concepts, with Llama3.2 demonstrating more gender-centric descriptions compared to Qwen2.5’s focus on objective physical attributes. These patterns may reveal underlying conceptual frameworks that shape how LLMs represent human subjects. Our analysis contributes to understanding representation in multimodal AI systems and has implications for reducing bias in descriptions of people.