What is MedGemma 1.5?

MedGemma 1.5 is an open AI model from Google that interprets medical text, 2D images, and 3D scans like CTs and MRIs, with enhanced accuracy for healthcare applications.

How does MedGemma 1.5 compare to MedGemma 1?

It adds support for high-dimensional imaging, improves accuracy by 3-35% on imaging tasks and up to 22% on text tasks, and is more efficient with a 4B parameter size.

What hardware do I need to run MedGemma 1.5?

For the 4B model, a GPU with at least 16GB VRAM (e.g., RTX 4090) is recommended; it can run offline, but larger variants need 32GB+ VRAM.

Is MedGemma 1.5 free and where can I get it?

Yes, it's free for research and commercial use, available on Hugging Face or Google Cloud Vertex AI, with tutorials on GitHub.

MedGemma

Open AI Model for Advanced Medical Imaging and Reasoning

From Google

Founder: Google Research Team

Data & Analytics

13 Jan 2026

1 Million+

Free

Pricing Model

Free

Starting Price

$0/Month

About This AI

MedGemma 1.5 is Google’s open multimodal AI model designed for healthcare developers, enabling interpretation of text, 2D images, and high-dimensional scans like CTs and MRIs.

With 4 billion parameters, it supports anatomical localization, longitudinal analysis, document extraction, and pairs with MedASR for accurate medical speech-to-text.

Analytics Image Editor LLM Transcript

Pricing

Pricing Model

Free

Starting Price

$0/Month

Key Features

Native support for 3D CT, MRI, and histopathology volumes
Anatomical localization in chest X-rays with high accuracy
Longitudinal imaging review for time-series data
Structured data extraction from medical documents
Integration with MedASR for low-error medical dictation

Pros

First open model for full 3D medical scan interpretation
Compute-efficient for offline use on modest hardware
Significant accuracy improvements over prior versions
Free and open-source with fine-tuning tutorials
Scalable on Google Cloud for production apps

Cons

Requires GPU for optimal performance
Limited to research/commercial use with validation needed
No built-in real-time processing for live scans
Dependency on quality of input data for best results

MedGemma is essential for healthcare developers building AI-driven diagnostic tools, offering unprecedented access to advanced multimodal analysis in an open framework.

FAQs

What is MedGemma 1.5?
MedGemma 1.5 is an open AI model from Google that interprets medical text, 2D images, and 3D scans like CTs and MRIs, with enhanced accuracy for healthcare applications.
How does MedGemma 1.5 compare to MedGemma 1?
It adds support for high-dimensional imaging, improves accuracy by 3-35% on imaging tasks and up to 22% on text tasks, and is more efficient with a 4B parameter size.
What hardware do I need to run MedGemma 1.5?
For the 4B model, a GPU with at least 16GB VRAM (e.g., RTX 4090) is recommended; it can run offline, but larger variants need 32GB+ VRAM.
Is MedGemma 1.5 free and where can I get it?
Yes, it’s free for research and commercial use, available on Hugging Face or Google Cloud Vertex AI, with tutorials on GitHub.