Medical Advisory Board: Dr. Ram Dandillaya, Clinical Chief, Cardiology, Cedars-Sinai; Dr. Antonio J. Forte, Faculty, Mayo Clinic; Dr. John J. Lee, Faculty, Harvard Medical School; Dr. Daniel Kahneman, Nobel Prize Laureate. OpenEvidence is a Mayo Clinic Platform Accelerate Company.

OpenEvidence: Your API to Medical AI

Simple and flexible pricing. Only pay for what you use.

First AI in history to score above 90% on the US Medical Licensing Exam

Developed as part of Mayo Clinic Platform Accelerate and powering Elsevier’s ClinicalKey AI, OpenEvidence is the only API to an AI purpose-built for the medical domain.

OpenEvidence scored 91% on the USMLE, ahead of Google Med-PaLM 2, Anthropic Claude 2, Chat GPT, and GPT-4.

Google Med-PaLM 2 Source: Google https://arxiv.org/pdf/2305.09617.pdf; GPT-4 Source: “Capabilities of GPT-4 on Medical Challenge Problems” https://arxiv.org/abs/2303.13375. ChatGPT Source: OpenAI. Claude 2 Source: Anthropic. Hippocratic AI’s self-reported value of 89.90% has been excluded since Hippocratic AI only took an unofficial, easier subset of USMLE practice questions; Hippocratic AI did not take (or at least report) scores for the official USMLE sample exam.

Mayo Clinic Platform
Elsevier
Harvard University
Forbes

OpenEvidence 2.0

Our most powerful model. Superior performance on tasks that require complex medical or scientific reasoning. The only AI system in history to score above 90% on the United States Medical Licensing Examination (USMLE), beating ChatGPT, GPT-4, Google Med-PaLM 2, and Anthropic Claude 2. With 128k context, always up-to-date medical knowledge and the broadest set of capabilities, OpenEvidence 2.0 is the most powerful medical AI on the market today.

Model Name

oe-2

Pricing

Contact for pricing details

OpenEvidence API is typically priced per medical analysis for a given topic or question. A medical analysis is an answer with a detailed clinical research overview and includes citations, sources, and references.