DeepSeek-R1 vs OpenAI o1 for Ophthalmic Diagnoses and Management Plans

Can DeepSeek-R1, an open-weight, reinforcement learning–enhanced large language model, outperform OpenAI’s proprietary o1 reasoning model in diagnosing and managing ophthalmology cases in clinical settings while reducing operating costs?In this cross-sectional evaluation of 422 cases from ’s Clinical Challenges, DeepSeek-R1 outperformed OpenAI o1 on diagnostic and next-step accuracy while reducing per-query costs.DeepSeek-R1 demonstrated superior performance and cost savings over proprietary models, highlighting its promise for scalable decision support in ophthalmology; however, clinicians should not rely on large language models for decisions, and results must be interpreted with extreme caution pending further validation in clinical settings.

Corresponding Author: Renaud Duval, MD, Department of Ophthalmology, Université de Montréal, 2900 Édouard Montpetit Boulevard, Montreal, QC H3T 1J4, Canada (renaud.duval@gmail.com).

Link to the article in your story

We encourage you to link out to this article in your story using the link below. It includes an access token that will give free access to the article for your readers up to one year after publication. (The link will be live after the article publishes and embargo is lifted.)

Please see the article for additional information, including full author list, author contributions and affiliations, conflict of interest and financial disclosures, and funding and support.

Need more information? Contact us.