← All models

MedGemma

Google · 2025 · Medical multimodal

Gemma 3 variants trained for medical text and images. Built for healthcare teams that cannot send patient data to the cloud. Research, documentation, and imaging literacy stay on the machine.

Maker
Google
Architecture
Medical multimodal
Fits from
Phone / edge · 6 GB+ RAM
Context
up to 33K
Runs
Local Intelligence
MedicalVisionClinical notesMedical imagesPrivate research

Sizes

MedGemma 1.5 4BQ4_K_M

requires 6 GB+ memory · 33K context · Phone / edge

2.5 GBWeights
MedGemma 4BQ4_K_M

requires 6 GB+ memory · 33K context · Phone / edge

2.5 GBWeights
MedGemma 27BQ4_K_M

requires 27 GB+ memory · 33K context · Workstation

17 GBWeights

Memory at context length

Approximate working set for MedGemma 1.5 4B as context grows.

4K
3.1 GB
8K
3.3 GB
16K
3.5 GB
33K
3.7 GB