
MedGemma 27B
Google's open medical LLM — 87.7% MedQA, built on Gemma 3.
About
MedGemma is Google's family of open medical AI models built on the Gemma 3 architecture, unveiled at Google I/O in May 2025. The text-only 27B variant targets medical reasoning and clinical Q&A, scoring 87.7% on MedQA (0-shot) and 89.8% best-of-5, while the 4B multimodal sibling handles radiology, pathology, dermatology, and ophthalmology images. The models are free to download and self-host via Hugging Face and Google's Model Garden.
Limitations: the weights ship under Google's Health AI Developer Foundations terms of use, not a standard open-source license — check commercial-use terms before deploying. It is a research-grade clinical assistant, not a certified medical device; outputs must be reviewed by qualified professionals.
Specs
Context window
Pricing
Free quota
API access
Weights
Released
Benchmarks
Provider-reported unless a source says otherwise. Benchmarks are marketing-adjacent by nature — treat them as directional.
Use cases
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 63; Trending score 82/100 from researcher assessment.