Peer Review of “The Performance of DeepSeek R1 and Gemini 3 in Complex Medical Scenarios: Comparative Study”

of DeepSeek R1 and Gemini 3 in Complex Medical Scenarios: Comparative Study." Round 1 Review General CommentsThis is a timely and well-structured paper [1] that investigates the application of DeepSeek R1, a state-of-the-art large reasoning model, in the medical domain using the Multitask Language Understanding Pro (MMLU-Pro) benchmark.This paper presents a follow-up evaluation of the DeepSeek R1 large reasoning model on open-ended medical scenarios from the MMLU-Pro benchmark.The study finds that DeepSeek R1 achieves a high accuracy of 92% without multiple-choice options, demonstrating its potential utility in more realistic clinical settings.The paper is timely and relevant, with strong empirical results and clear motivation.However, it would benefit from revisions to improve clarity, contextual grounding in existing work, and methodological detail.The authors may also consider citing recent work that examines the questioning strategies of large language models (LLMs) in clinical dialogues to better position this study in the broader landscape.