A free and trustworthy platform for comparing LLMs on medical queries
To begin, please sign in:
By using this website, you agree to our Terms of Use.
Privacy: Usage and data are anonymized. No personal information is stored other than email address or ID for sign in. This research study has undergone IRB review and exemption.
MedArena is a free platform for clinicians to compare different large language models (LLMs) in an unbiased, head-to-head competition.
Submit a medical query, and MedArena will present responses from two randomly chosen models. Choose the superior response, and your input will help refine the models over time.
MedArena was developed by the Zou Lab at Stanford University. We are grateful for our collaborations with NEJM AI, Doximity, Centaur Labs, and Eric Topol. We also extend our thanks to LMSYS.org for their inspiration.