How much does OpenEvidence cost?

Osler, Sackett and Snow are free for clinicians whose medical licence OpenEvidence has verified, on openevidence.com and the iOS and Android apps. There is no paid tier for these three models. Darwin, the fourth model, is not sold at all — access is granted by application to institutional partners and academic researchers.

Can I get access to OpenEvidence Darwin?

Darwin is in research preview and available by application only. OpenEvidence grants access to institutional partners such as the National Organization for Rare Disorders, research collaborators, and accredited AI researchers at academic institutions. The company points to dual-use risk in virology and genetics work as the reason Darwin is not open to every verified clinician.

How does Darwin compare to other medical AI models?

OpenEvidence reports Darwin at 72.8% on MedXpertQA against 65.1% for the next-best baseline, 82.7% on HealthBench Professional against 70.6%, and a severity-weighted F1 of 0.872 on NOHARM against 0.740. Darwin also answers all 660 MedQA questions correctly, which OpenEvidence calls the first perfect score on that benchmark.

Does OpenEvidence handle cancer care?

OpenEvidence pairs a specialised oncology sub-agent with a precision oncology knowledge base added through a partnership with a leading cancer centre. The platform already draws on NCCN treatment algorithms and ASCO guidelines, so oncology questions are answered against those sources rather than general web search.

How many clinicians use OpenEvidence?

OpenEvidence reported 1.12 million medical licence-verified U.S. clinicians on the platform as of September 2026. That figure counts verified accounts rather than daily users, and it is the base the new Osler, Sackett and Snow models roll out to at no cost.

OpenEvidence · 2026-09-03 · major

OpenEvidence model family — Osler, Sackett and Snow ship free to clinicians

OpenEvidence released four medical AI models named after figures in medical history. Osler, Sackett and Snow are free for license-verified clinicians. A fourth model, Darwin, scores a perfect 660/660 on MedQA and is research preview only.

OpenEvidence medical AI model family launch with Darwin research preview — Unite.AI

Four medical AI models that all aim at the same accuracy — what changes is how long each one thinks.

Key specs

Med qa (darwin)	660/660
Verified us clinicians	1.12M

Quick facts

Maker	OpenEvidence
Models	Osler, Sackett, Snow (live); Darwin (research preview)
Fastest	Osler, about 5 seconds per answer
Deepest	Snow, about 5 minutes per answer
Price	Free for license-verified clinicians
Availability	openevidence.com plus iOS and Android apps
Darwin access	By application only

Benchmarks

MedXpertQA

OpenEvidence Darwin		72.8%
Next-best baseline		65.1%

source ↗

HealthBench Professional

OpenEvidence Darwin		82.7%
Next-best baseline		70.6%

source ↗

What is it?

OpenEvidence now ships a named model family instead of one search box. Osler answers in about five seconds and is the new default. Sackett takes about thirty seconds for questions that turn on how strong the evidence is. Snow takes about five minutes and reads through the medical literature before it writes anything.

How does it work?

Every model in the OpenEvidence family is held to the same standard of clinical accuracy, so the difference between them is time — how long the model thinks and how deep it searches. Snow replaces the older Deep Consult feature and runs a full literature investigation before producing a report. A separate oncology sub-agent draws on a precision oncology knowledge base alongside NCCN treatment algorithms and ASCO guidelines.

Why does it matter?

A clinician can now match the wait to the question instead of getting one fixed speed. OpenEvidence says 1.12 million license-verified U.S. clinicians use the platform, and Osler, Sackett and Snow cost them nothing. The fourth model, Darwin, is held back to research preview because OpenEvidence points to dual-use risk in virology and genetics work.

Who is it for?

practising clinicians and clinical AI researchers

Frequently asked questions

How much does OpenEvidence cost?: Osler, Sackett and Snow are free for clinicians whose medical licence OpenEvidence has verified, on openevidence.com and the iOS and Android apps. There is no paid tier for these three models. Darwin, the fourth model, is not sold at all — access is granted by application to institutional partners and academic researchers.
Can I get access to OpenEvidence Darwin?: Darwin is in research preview and available by application only. OpenEvidence grants access to institutional partners such as the National Organization for Rare Disorders, research collaborators, and accredited AI researchers at academic institutions. The company points to dual-use risk in virology and genetics work as the reason Darwin is not open to every verified clinician.
How does Darwin compare to other medical AI models?: OpenEvidence reports Darwin at 72.8% on MedXpertQA against 65.1% for the next-best baseline, 82.7% on HealthBench Professional against 70.6%, and a severity-weighted F1 of 0.872 on NOHARM against 0.740. Darwin also answers all 660 MedQA questions correctly, which OpenEvidence calls the first perfect score on that benchmark.
Does OpenEvidence handle cancer care?: OpenEvidence pairs a specialised oncology sub-agent with a precision oncology knowledge base added through a partnership with a leading cancer centre. The platform already draws on NCCN treatment algorithms and ASCO guidelines, so oncology questions are answered against those sources rather than general web search.
How many clinicians use OpenEvidence?: OpenEvidence reported 1.12 million medical licence-verified U.S. clinicians on the platform as of September 2026. That figure counts verified accounts rather than daily users, and it is the base the new Osler, Sackett and Snow models roll out to at no cost.

Try it

https://www.openevidence.com