Long questionnaires are burdensome to respondents, while open questions allow for greater elaboration and nuance. Large Language Models (LLMs) offer a scalable way to evaluate text. Prior work has examined LLM-based personality assessment for select traits; this study tests whether LLM-generated scores from a single open question converge with NEO-FFI scores across all Big Five traits, and whether the open-question format is more time-efficient. Two hundred participants completed an online survey consisting of the NEO-FFI and a single open question. Response times were tracked and an LLM was used to generate NEO-FFI scores based on the open text. Quantitative and qualitative post-hoc analyses were conducted to explore the mechanisms driving convergence. Correlations between NEO-FFI and LLM scores were significant, and on average positive and moderately large for all personality traits. Factors such as trait and the LLM’s self-evaluated certainty impacted convergence. Contrary to expectations, the open question was not more time-efficient than the NEO-FFI. Although LLM-based personality scoring from open text shows promise, practical benefits remain unclear, and notable limitations warrant further investigation before applying this approach in practice.