
by OpenEvidence
OpenEvidence is a medical AI platform for clinicians that gives citation-backed answers drawn from peer-reviewed literature at the point of care. It is free for verified clinicians.
Web appFreemium
Step 1: Sign in as a verified clinician.
Step 2: Pick a model from the selector; start with the default for fast answers.
Step 3: Type a clinical question in the ask bar, such as a dose, diagnostic criterion or first-line therapy.
Step 4: Review the answer and its citations, and move to a more thorough model if the question needs more nuance.
Step 5: Use Visits, Voices or Dotflows to document encounters and shape how responses and notes are written.
The pages advise starting with the default model for most questions. Use the middle model when you want a broader survey or a clarifying question, and the deepest one for hard cases when you can wait several minutes for a report.
Dotflows are reusable natural language prompts that customize responses. You trigger them by typing a period in the ask bar, and you can build your own or browse a community library.
It is a research preview available by application only, for institutional partners, research collaborators and academic researchers benchmarking AI accuracy and safety in clinical medicine.
OpenEvidence says answers are grounded in peer-reviewed research and that it has content agreements with NEJM, JAMA, Nature, NCCN and Cochrane, among others.